THESiS

\D
)4
\

This is to certify that the
dissertation entitled

TRUTH VS BEAUTY:
ESSAYS 0N STANDARDS, TRADE, AND AGREEMENTS

presented by
MONIKA TOTHOVA

has been accepted towards fulfillment
of the requirements for the

Doctoral degree in Agricultural Economics and
Economics

Kiwi/f"? /WE Sig/LH/Mtlyéf

Majo/ Professor’s Signature

&M#[{ 22 Z mg élfigzgjl Q? @004

Date

 

 

MSU is an Affinnative Action/Equal Opportunity Institution

 

LIBRARY
Michigan State
University

 

 

 

 

TRUTH vs BEAUTY:
ESSAYS ON STANDARDS, TRADE, AND AGREEMENTS
By

Monika Tothova

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Agricultural Economics
Department of Economics

2004

ABSTRACT
TRUTH vs BEAUTY: ESSAYS ON STANDARDS, TRADE, AND AGREEMENTS
By

Monika Tothova

As the WTO succeeded in reducing tariffs, non-tariff trade barriers (N TBS) seem
to be on the rise. Among the NTBs, the premium position is occupied by discrepancies
across countries’ standards. Countries are guaranteed the right to choose standards they
deem appropriate for human, animal, and plant health, assuming they do not serve as
barriers to trade. This dissertation examines the role of standards in trade: it investigates
interactions among countries with different standards and preferences. The borderline
cases of standards considered are “Truth” (defined as the good having a certain attribute)
and “Beauty” (defined as the good missing a certain attribute). From an individual
country’s perspective, standards are perceived as vertically differentiated (that is, the
Home consumer feels Truth is superior to Beauty, but the Foreign consumer considers it
inferior). However, from the global perspective, standards are not directly comparable in
terms of their performance, and according to the WTO, goods that satisfy differing
standards may be classified as “like products”. The dissertation consists of three related
essays and follows a traditional theoretical model-building track with an application to
genetically modified foodstuffs.

The first essay, Standards as Strategic Policy Instruments, considers a problem
where two pure exchange economies with different preferences each select standards

while also trying to accommodate both domestic and foreign consumers (who have

different preferences) in order to obtain gains from terms of trade effects. Four different
steady states are explored and compared: (1) the social optimum, (2) myopic equilibrium,
(3) a Nash equilibrium game, and (4) a Nash bargaining solution.

The second essay, Differinétaidards. Welfare, and Sub-global Agreements.
explores endogenous standards-setting in a Krugman-like model of monopolistic
competition to investigate formation of potential trade agreements when countries do not
have an opportunity to compromise on their standards. Gains from trade arise from a
larger number of available varieties. When a global agreement on prevailing standards (or
any other NTB) is not feasible, countries resume forming trade agreements with partners
that have similar preferences as an alternative way to secure some gains from trade.

In the third essay, Economies of Scale: Autarky, Harmonization, and
Compromise, standards are modeled as tolerance levels. Gains from trade come from
larger numbers of varieties available, as well as traditional economies of scale. In an open
trade environment we study the impact of standards harmonization and compromise on
the social welfare of each country, and conclude that unless the economies of scale are
large enough to justify adjustments, countries are better off in autarky.

The essays are complemented with a case study of genetically modified foodstuffs
(in reality the EU position lies close to the “Truth” boundary, while the North American
position lies close to the “Beauty” boundary) and explores the policy struggles introduced
by their commercial approval in 1992. Countries’ positions suggest potential gains from
trade are not sufficient to justify harmonization, and countries cluster into clubs based on

similar preferences and actual or perceived gains from trade.

Copyright by
MONIKA TOTHOVA

2004

DEDICATION

To the memory of my grandparents:

Pavel Gyurka, Mici K6berling, and Maria Stredova

ACKNOWLEDGMENTS

It is undeniably a triumph for all parties involved to see the grand project of my
formal education finally coming to realization. However none of it would have been
possible without the support and guidance of many people who I would like to thank.

I would like to thank the members of my dissertation committee: Steve Matusz,
James Oehmke, David Schweikhardt, and Jay Wilson. They all contributed in different
but significant ways to my education. Steve’s door was always open and his remarks
never limited to the dissertation itself. I received his guidance regarding the job market,
research papers, and career choices — often on less than a short notice. Jim, who on many
occasions demonstrated he indeed is a man of patience, bestowed his confidence in me to
undertake such a project. Thanks to him I was privileged to travel, see, and be seen. As
my major professor, Dave has been supportive throughout my graduate studies. Jay
pushed me to think harder in my theory and ask the right questions.

My studies and travels at Michigan State would not have been possible without
the financial support and generosity of many institutions and individuals. Major funding
from the Elton R. Smith Endowment, Thoman Foundation, and Glenn and Sandy Johnson
Fellowship is gratefully acknowledged. Dr. Sandra Batie — the holder of the Elton R.
Smith Professorship in Food and Agricultural Policy — has always been generous with
providing me funding whenever the need arose. Thomas Jeitschko did an excellent job in
preparing me for the job market and providing moral support during the process.

I have benefited immensely from the enthusiastic support of Willi Meyers, now at

University of Missouri in Columbia. Willi has kept my spirits high in times good and

vi

bad, given me boosts of confidence when I needed it most and has worked hard so that
the value of my professional network is maximized. Louise Sause became a dear fiiend:
thanks go to her for keeping a watchful eye over me at all times and numerous meals
shared. Thanks to my aunt Helena Kucerkova for her positive outlook.

My life here in East Lansing would have been much more challenging and dry
had it not been for the fi'iendship of the “gang”: Irina Agafonova, Artem Prokhorov and
Vlado Hlasny. Together we formed a (dys)functional family of sorts sharing good food,
wine and gossip, that melted together to form support and understanding. Thanks to
Cecilia Samonte for shopping trips and her camaraderie during the days spent in the
library while studying for our prelirns, Miro Student for making me cook every Friday,
and Juan Manuel Estrada-Valle for ensuring that I had a social life. Kellie Raper and
Cynthia Donovan provided me a place to stay during times of temporary dislocations, for
which I am deeply grateful.

Thanks to Anwar Naseem for his superior handling of my troubled personality -
especially when numerous crises struck. His kindness, support, sense of humor, and
ability to evaluate life’s small disasters from a distance, surely brought a lot of joy and
calm. I could have not wished for a more understanding bloke.

Finally I thank my parents Hilda and Imrich who did not realize the consequences
of their rearing emphasizing the need for higher education until both their children ended
up with advanced degrees. Thanks for their love and sacrifices they made to let me
achieve my dreams. Thanks to my brother Pavol “Jonathan” for his love - and for staying

at home.

vii

TABLE OF CONTENTS

LIST OF TABLES .............................................................................................................. x
LIST OF FIGURES ........................................................................................................... xi
CHAPTER 1: INTRODUCTION ....................................................................................... 1
1.1. Standards ............................................................................................................. 3
1.1.1. Institutional Background ............................................................................. 3
1.1.2. Standards Setting within the WTO Framework .......................................... 5
1.1.2.1. Scientific Basis and the Precautionary Principle ....................................... 7
1.1.2.2. Conformity Assessment Procedures .......................................................... 8
1.1.2 Literature: Standards and Product Differentiation in Trade ....................... 9
1.2. Trade Agreements ............................................................................................. 11
1.2.1. Institutional Background ........................................................................... 11
1.2.2. Literature: Trade Agreements ................................................................... 12
1.3. Standards and Trade Integration ....................................................................... 14
1.3.1. Anecdotal Evidence: Standards (N TBS) and Agreements ........................ 14
1.3.2. Literature: Standards and Coalitions ......................................................... 17
1.4. Contribution to the Literature ........................................................................... 19
CHAPTER 2: STANDARDS AS STRATEGIC POLICY INSTRUMENTS .................. 21
2.1. Model Description ............................................................................................ 22
2.2. Case 1: Social Planner ....................................................................................... 27
2.3. Case II: Myopic Equilibrium ............................................................................ 31
2.4. Equilibrium Price .............................................................................................. 34
2.5. Case 111: Nash Game ......................................................................................... 36
2.4.1. Nash Equilibrium with Identical Endowments ......................................... 40
2.4.2. Nash Equilibrium with Different Endowments ........................................ 43
2.6. Case IV: Nash Bargaining Solution .................................................................. 48
2.7. Chapter Conclusions ......................................................................................... 50
CHAPTER 3: DIFFERING STANDARDS, WELFARE, AND SUB-GLOBAL
AGREEMENTS ................................................................................................................ 52
3.1. General Model Setup ........................................................................................ 52
3.2. Imposed Harmonization Potentially Causing Loss of Welfare ......................... 64
3.2.1. Autarky ..................................................................................................... 65
3.2.2. Unilateral Harmonization .......................................................................... 67
3.2.2.1. Large vs. Small Country Harmonization ................................................. 75
3.2.3. Non-harmonized Trade ............................................................................. 78
3.3. Large Nmnber of Countries .............................................................................. 80
3.3.1. Non-standardized Trade between Clubs ................................................... 82
3.3.2. World Standardized Trade between Clubs ............................................... 84
3.4. Core Analysis .................................................................................................... 87
3.5. Applying Lessons from the Club Theory .......................................................... 88
3.6. Chapter Conclusions ......................................................................................... 91
CHAPTER 4: ECONOMIES OF SCALE, AUTARKY, HARMONIZATION, AND
COMPROMISE ................................................................................................................ 94
4.1. Consumption ..................................................................................................... 96
4.2. Production ......................................................................................................... 99

viii

4.3. Competitive Equilibrium vs. Social Optimum ................................................ 101

4.4. Autarky Decision ............................................................................................ 102
4.5. Open Trade ...................................................................................................... 104
4.5.1. Unilateral Harmonization ........................................................................ 106
4.5.2. Mutual Compromise ............................................................................... 109
4.5.3. No Trade Occurring due to Large Differences in Standards .................. 111
4.6. Two Countries Case with Equal Size Assumption Released .......................... 111
4.7. Chapter Conclusions ....................................................................................... 113
CHAPTER 5: CONCLUSIONS AND BIOTECHNOLOGY APPLICATION ............. 114
5.1. Conclusions from the Theory Essays .............................................................. 114
5.2. GM Backgrounder .......................................................................................... 118
5.2.1. Economic Importance ............................................................................. 119
5.2.2. Reasons for Controversy and Technology Regulation ........................... 120
5.3. Controversy: EU vs. US .................................................................................. 123
5.4. Model Application .......................................................................................... 125
5.5. Anecdotal Evidence from Third Countries: Small vs. Large .......................... 128
5.6. Deeper Conflict: Countries vs. Multilateral Agreements ............................... 130
5.7. Implication for the World Trading System ..................................................... 135
5.8. Chapter Conclusions ....................................................................................... 140
Appendix A: Standards as Strategic Policy Instruments ............................................ 144
A. 1. Social Planner Problem with Identical Endowments .................................. 144
A2. Social Planner Problem with Different Endowments ................................. 147
A3. Myopic equilibrium, Second order conditions ............................................ 152
AA. Price Behavior ............................................................................................. 154
A. 5. Nash Equilibrium ........................................................................................ 156
A. 6. Nash Bargaining Solution ........................................................................... 158
Appendix B: Differing Standards, Welfare, and Sub-global Agreements .................. 162

B. 1. Tradeoff between Number of Varieties and Price: “Quantity and Quality” 163
B2. Existence of Equilibrium in Open Trade with No Harmonization Required
169
BIBLIOGRAPHY ........................................................................................................... 1 75

ix

LIST OF TABLES

Table 2.1: Location of a Nash Equilibrium Candidate: Home Country ................ 39
Table 2.2: Location of a Nash Equilibrium Candidate: Foreign Country ............... 40

Table A.1: Social Planner: Numeric Solutions, Stationary Points, Identical
Endowments .................................................................................... 144

Table A2: Social Planner: Numeric Solutions, Stationary Points, Different

Endowments .................................................................................... 146
Table A3: Sign of the Second Derivative of the Social Welfare Function ............ 151
Table A.4: Simulations of Nash Equilibrium with Identical Endowments ............ 156
Table A5: Simulations of Nash Equilibrium with Different Endowments ............ 157

Table A6: Simulations of Nash Bargaining Solution with Identical Endowments 158

Table A.7: Simulations of Nash Bargaining Solution with Different Endowments .. 160

Table 8.1: Production — Entire Stock of Labor Directed into a Single Quality ........ 162
Table 3.2: Autarky Prevailing Standards in Each Country .............................. 167
Table 83: Numbers of Varieties in Autarky and Unilateral Harmonization ......... 168

LIST OF FIGURES

Figure 2.1: Behavior of a Exponents as a Function of h .................................. 25
Figure 2.2: Behavior of 6 Exponents as a Function off. ................................. 25
Figure 2.3A: Home: Indirect Utility Function without Terms of Trade Effects ....... 32
Figure 2.3B: Foreign: Indirect Utility Function without Terms of Trade Effects. . . .. 33
Figure 2.4A: Price as a function of h ........................................................ 35
Figure 2.48: Price as a Function off. ....................................................... 36
Figure 2.5: Nash Equilibrium with Country-variant endowments ....................... 43
Figure 2.6: Nash Bargaining Solution, Country-variant Endowments .................. 50

Figure 3.1: PPF in the Standard-variant Marginal, Constant Fixed Labor
Requirement Case .............................................................................. 61

Figure 3.2: Welfare-maximizing Choice of Standard in Standard-variant MC Case. 62

Figure 3.3: Social Welfare as a Function of A: Autarky ................................... 66
Figure 3.4: Social Welfare as a Function of A: Autarky, Standardization to Truth
and Beauty, from the Home Country Perspective .......................................... 69
Figure 3.5: Social Welfare as a Function of i: Autarky, Standardization to Truth
and Beauty, from the Foreign Country Perspective ....................................... 71
Figure 3.6: Welfare Ranking of Alternatives ............................................... 73

Figure 3.7: Agreements and Prevailing Standards, Case of Identical Endowments... 74

Figure 3.8: Social Welfare as a Function of A: Autarky, Standardization to Truth
and Beauty. Differently Sized Countries, Home Modeled as Small ..................... 76

Figure 3.9: Prevailing Standard: Home Modeled as Small, Foreign as Large
Country ......................................................................................... 77

Figure 3.10: Social Welfare as a Function of A: Autarky, Non-standardized Trade,
6 > 0.5 79

xi

Figure 3.11: Social Welfare as a Function of ,1: Autarky, Non-standardized

trade, 0 < 0-5 .................................................................................... 79
Figure 3.12: Social Welfare as a Function of A: Autarky, Club .......................... 82
Figure 3.13: Social Welfare as a Function of 1.: (t+b) Countries, Autarky, Non-
standardized Trade, 9 > 0-5 .................................................................. 83
Figure 3.14: Social Welfare as a Function of A: (t+b) Countries, Autarky, Non-
standardized Trade, 6 < 0'5 .................................................................. 34
Figure 3.15: Social Welfare as a Function of ,1: (t+b) Countries, Autarky,

Standardized Trade ............................................................................. 85
Figure 4.1: Autarky Decision in General Equilibrium .................................... 103
Figure 4.2: Number of Varieties ............................................................. 105
Figure 4.3: Decision to Harmonize ......................................................... 107
Figure 4.4: Decision to Compromise: The Foreign Perspective ......................... 110
Figure 4.5: Small vs. Large Country ......................................................... 112

Figure A.2: Social Welfare as a Function of Choice Parameters h, f, Identical
Endowments, Difference between Preference Parameters “small” ...................... 146

Figure A.2: Social Welfare as a Function of Choice Parameters h, f, Identical
Endowments, Difference between Preference Parameters “large” ...................... 146

xii

CHAPTER 1

INTRODUCTION

A sequence of struggles has accompanied global trade institutions ever since the
earliest failed attempt to establish a world trade organization following the Second World
War. While the jury is out on causality — whether lack of progress on the global level has
been pushing countries into smaller arrangements, or efforts spent on the smaller
agreements are inhibiting global advancement - sub-globall agreements are on the rise.
Between 1948 and 1994, 124 trade agreements were signed among the contracting parties
to the General Agreement on Tariffs and Trade (GATT). Since its creation in 1995, the
World Trade Organization (WTO) was notified of over 130 different agreements
covering goods and services.

Even as the GATT/WTO succeeded in reducing tariffs in most sectors, it has been
faced with disputes related to non-tariff trade barriers (NTBs) — for example not yet fully
addressed processes standards often resulting in production of “like” productsz. In
addition to the dispute over “dolphin-safe” tuna, among the most notorious cases are the
gasoline environmental standards (brought by Venezuela against the US), meat hormones
(the US vs. the EU), Shrimp vs. turtle (India, Malaysia, Pakistan, Thailand vs. the US),
etc. While believers in market forces claim that demand makes grades and standards
redundant, in cases of indistinguishable like products some regulation has to be in place

in order to facilitate the consumer’s decision. Article 20 of the GATT allows

 

' Terminology: sub-global (also labeled as regional or preferential) trade agreements include arrangements
between two and more countries but not all countries necessarily being within a regional proximity.
2 Like products are defined as products with the same end use and identical tariff classification.

1

governments to act on trade to protect human, animal or plant life or health, provided
they do not discriminate or use this as disguised protectionism. Countries’ rights to adopt
the standards they consider appropriate is also recognized by the two specific WTO
agreements dealing with food safety and animal and plant health and safety (Agreement
on Sanitary and Phytosanitary Measures), and with product standards (Agreement on
Technical Barriers to Trade), assuming their use is justified and they are not used as
barriers to trade (WTO, 2004). Thus, countries are allowed to set and require different
standards as long as imported and domestic goods are treated equally.

This dissertation explores the relationship between countries’ differences in
standards, which are rooted in their preferences, and the increasing numbers of small
trade agreements. It investigates the emergence of different standards across countries;
the impact of standards altered to capture gains from trade on individual countries’
welfare, and in some cases extends to the formation of trade agreements. Chapter Two
considers a case of pure exchange economies and explores the differences of standards
selection in different settings, allowing for different types of extemalities imposed by a
choice of standard. Chapters Three and Four develop a Krugman-esque model of
monopolistic competition and trade. The model is amended so clear consumer preference
leads to the endogenous formation of national standards. Our framework can account for
both goods and processes standards — as long as they impact the cost of production with
internal scale economies. Economies of scale and gains from trade (larger number of
varieties available) motivate the endogenous formation of trade agreements, which may
require modification of national standards. A global agreement can be formed on the

basis of a uniform international standard or harmonization of national standards. In the

absence of concord on prevailing standards, countries cluster into sub-global agreements
based on their preferences. The dissertation is complemented by a case study of
biotechnology and genetically modified foodstuffs — an issue not yet fully attended to on
the global level, and intensively stirring the trade relations between the EU and the US.

There appear to be two main reasons for harmonization of standards to occur in
trade: (1) to synchronize upstream and downstream production if intermediate goods are
traded or (2) to remove non-tariff (or technical) barriers to trade for end-use products, i.e.
when a standard in question does not induce synchronization in the production. The
dissertation considers the second case.

The preface introduces the institutional background and literature behind
standards3 and sub-global agreements. Descriptions, solutions and preliminary
conclusions of different models are presented in chapters Two, Three, and Four. Chapter
Five summarizes the findings of the theoretical essays, and applies them to the
controversy on transgenetic crops and genetically modified foods. Appendices follow

Chapter Five.

1.1. Standards

1 .1 . l . Institutional Background

Standards are the product of market forces or government intervention (Bhalla
and Kennedy, 1998). In trade law a standard is a document approved by a recognized

body (either governmental or non-govemmental, e.g., the International Organization for

 

3 We use standards as an example of non-tariff trade barrier (NTB) in its broadest sense. Indeed, the model
can be extended to account for any other NTB, such as labeling, production process, etc.

3

Standardization"), that provides for common and repeated use, rules, guidelines or
characteristics for products or related processes and production methods, with which
compliance is not mandatory. A technical regulation is a document which lays down
product characteristics or their related production processes; compliance is mandatory. A
technical regulation is usually government imposed (Bhalla and Kennedy, 1998).
Following the literature, we use the term “standard” to encompass both types of
documents.

Wilson (2001) categorizes standards by function into product and process
standards. Product standards refer to characteristics that goods should possess — for
example, size, minimum nutrition content, maximum pesticide residue on agricultural
products, etc. Process standards (ofien labeled as production processes methods —- PPMs)
refer to conditions under which products are manufactured, packaged, or refineds . PPMs
may be targeted at production conditions that are not directly related to the final good
(Wilson, 2001), such as labor and environmental standards. The GATT recognizes the
concept of “like products” based on tariff classification and end use: if products fall into
the same tariff classification category and are subject to the same end use, they are to be
treated identically, and countries cannot discriminate on the basis of the PPM.

Standards are implemented as a solution to an asymmetric information problem
contained in the goods whose attributes are known to the producer only. While

potentially costly to launch, standards — once established - are non-excludable and non-

 

’ The lntemational Organization for Standardization (ISO) is a network of the national standards institutes
of 146 countries, on the basis of one member per country, headquartered in Geneva. ISO standards are
voluntary, as a non-govemmental organization, ISO has no legal authority to enforce their implementation.
Most of the ISO standards are technical, i.e., methods standards (http://www.iso.ch. last viewed on June 1,
2004).

5 Examples of PPMs are meat radiation, hormone treatment, tomato ripening, “rainforest-safe bananas”,
“sweatshop-free textiles”, “leg hold trap-free fur”, “fair trade coffee”, “sustainable harvested lumbef’or
“dolphin-safe tuna”.

depletable, and exhibit characteristics of public goods. Due to their national character
standards are often perceived as a domestic policy instrument imposed, to facilitate
market exchange, smooth upstream and downstream interactions among industries, lower
transaction costs among agents as a medium of information exchange, and utilize

economies of scaleé.

1.1.2. Standards Setting within the WTO Framework

With the progressive elimination of tariff barriers, the debate on international
trade policies has shifted to national differences on standards and regulations (Casella,
2001). In spite of the positives standards introduce, and the countries’ rights to limit the
entry of the products to the markets, these limitations are non-tariff barriers to trade
providing economic benefits to domestic firms that produce import-competing goods
(Kerr, 2003). Harmonization of standards in the manufacturing sector has been less
troublesome than harmonization in agricultural and food sectors — perhaps due to larger
perceived gains from harmonization in the industrial sector. Standards that differ between
countries have the potential seriously to impede trade (Sampson, 2000) —- as currently

reflected by the example of US. corn exports to the EU7. After the Uruguay Round it was

 

6 With the increasing realization of network extemalities, standards acquired a new role as a network
facilitating instrument. Network extemalities arise when the value of the network depends on the number of
users in the network. Standards assist in exploring extemalities in a single interconnected network, and
open standards lessen the probability of lock-in (the potentially substantial cost of switching a technology
once an investment in a different technology has been made, i.e., PC and Mac) (Shapiro and Varian, 1999).
Some hold standards responsible for a potential loss of varieties (ditto). A repeated argument challenging
standardization often cites a case of the QWERTY keyboard as being locked in an inferior technology (i.e.,
Shapiro and Varian, 1999; David, 1985). We pay attention only to the internal scale economies, and ignore
probable network effects. We also overlook the details of institutional aspects of standard setting.

“U.S. farmers estimate the European restrictions [on genetically modified foods] have cost them nearly

$300 million a year in lost corn exports alone.” (Ligp://www.cnn.com/2003/WORLD/europe/07/02/eu.gr_n.
crops/, last viewed on June 7, 2004). Under the rules passed by the European Parliament in July 2003,

5

hoped that harmonized international standards would evolve by the WTO recognizing
three science-based institutions as having competence to develop international standards:
(1) the Codex Alimentarius for food safety; (2) the International Office of Epizootics
(animal health); and (3) the Secretariat for International Plant Protection Convention
(plant health) (Kerr, 2003). Currently standards are becoming more widespread, but
remain disordered. For example, tolerance levels for tetracycline in beef vary from 0.1
parts per million (ppm) in the EU and New Zealand, through roughly 0.2 ppm in Japan,
Canada and Australia to 2.0 ppm in the United States. The level recommended by the
Codex Alimentarius is 0.6 ppm. Wilson et a1 (2002) estimate that harmonization at the
Codex standard would increase bovine meat trade by $3.2 billion (or 57%), and that the
trade value at the Codex standard is $5.1 billion higher than the trade value under the
most stringent standard (0.1ppm).

The Agreement on Technical Barriers to Trade (TBT) — a specific WTO
agreement balances the ability of governments and the private sector to implement
legitimate standards and the procedures for assessing product conformity with those
standards against their unjustified use to protect a domestic industry (Bhalla and
Kennedy, 1998). It does not cover services, sanitary and phytosanitary measures or
governmental purchasing specifications. It does cover regulations and standards on PPMs
to the extent that they relate to product standards characteristics, but not as they relate to
pollution or extemalities caused by PPMs (Bhalla and Kennedy, 1998). Article 2.1 of the
TBT Agreement states that “Members shall ensure that in respect to technical regulations,

products imported from the territory of any [WTO] member shall be accorded treatment

 

products with more than 0.9 percent of biotech material would have to be labeled with the words: “This
product is produced from GMOs”. No similar legislation is in place in the US.

6

no less favorable than that accorded to like products (i.e., with the same end use and tariff
classification) originating on any other country” (italics added).

The Agreement on Sanitary and Phytosanitary Measures (SPS) “allows countries
[to] set their own standards assuming they are based on science to the extent necessary to
protect human, animal or plant health and not arbitrarily or unjustifiably discriminate
between countries where identical or similar conditions prevail” (WTO, 2003). Intense

disputes exist on what are the appropriate standards to apply for the protection of health.

1.1.2. 1. Scientific Basis and the Precautionary Principle

The concept of science-based rules for the establishment of trade barriers is “an
attempt to de-politicize decision making in the complex areas of human, animal, and
plant health as well as aspects relating to the environment” (Kerr, 2003). A crucial
question of risk management (a scientific determination of the relationship between cause
and effects in situations where adverse effects can occur) to human, animal, and plant life
and health is deciding on the acceptable risk levels and the appropriate standards to
adopt. It raises an infamous question of the relationship between risk and precaution. The
Precautionary Principle states that where there are threats of serious or irreversible
damage, lack of full scientific certainty shall not be used as a reason for postponing cost-
effective measures to prevent the eventual outcome of risk, covering the gap between
banning a product or procedure until science has proved it is harmless and not banning it
until science has proved that there is a real risk (Sampson, 2000).

While there is general agreement on exercising precaution in the face of

uncertainty, there is no agreement on how it should be operationalized for decision

making (Kerr, 2003). The most politicized is the debate over genetically modified (GM)
products: US. regulators view them as not substantially different from the conventional
variety, while the EU adheres to the “precautionary principle” and sees GM and

conventional as differentiated products due to the perceived risks (Nielsen et al., 2002).

1.1 .2.2. Conformity Assessment Procedures

While standards themselves undeniably are central in shaping trade relations, a
subsequent problem is mutual recognition of similar standards as well as related
conformity assessment procedures (CAP). Article 5.1.1 and 5.2.1 of the TBT Agreement
provides for the MFN treatment on CAP. Countries — although discouraged from doing
so - can choose to deny entry of foreign goods unless their conformity is reassessed (i.e.,
labeled with a consumer information label to fit domestic requirements). Obligation to
comply with foreign standards requirements could result in increased cost for foreign
producers, and includes all aspects of CAP, including laboratory accreditation and quality
system registration. The minimum derogation principle (Art. 5.12) states that CAP “shall
not be more strict or be applied more strictly than is necessary to give the importing
member adequate confidence that products conform with the applicable technical
regulations or standards taking into account the risk non-conformity would create”.
Similarly, countries are encouraged to recognize practices — which, albeit different —
yield the same result, for example in testing for qualitative and quantitative
characteristics. For the purposes of this paper we do not allow countries to mutually

recognize their CAPS, and focus on standards per se.

1.1.2 Literature: Standards and Product Differentiation in Trade

The dissertation addresses standards regarding like products. Therefore, the usual
reasons for standardized production to ensure compatibility of the upstream and
downstream production processes are not relevant.

Standards in literature are frequently treated as reflections of domestic policies
and preferences (Ederington etal., 2002) most frequently imposed for facilitation of
market exchange. With decreasing tariff trade barriers, it is suspected that standards are
also employed as a form of strategic trade policy (e.g., Dasgupta, 2000; Mattoo, 2001).
Most of the up to date literature on standards and trade has dealt primarily with process
standards, i.e. environmental and labor standards (e.g., Lynch, 2000; Griffin, 2000, etc);
and standards in the context of development — such as exploring the impact of standards
imposed by developed countries on their developing countries’ suppliers, often on the
farm level (Henson, 1999; Donovan et al., 2001; Hufbauer et al., 2002).

Not every paper is concerned with standards imposed as means of trade policy;
rather, some investigate social dimensions of standards and their impacts on domestic
markets.

The most comprehensive theoretical model of product standards in international
trade is Casella (2001), explicitly recognizing the role of government in establishing
general guidelines for standards. Contrary to the usual treatment considering standards a
public good, Casella explicitly models standards as club goods while assuming a pure
exchange economy, and accounting for the provision of the club goods as well. Clubs
originate in private coalitions of firms as developers of exact specifications, and thus her

model investigates the mechanics of potential harmonization from the bottom. The size of

national standard-sharing coalitions is determined by the trade-off between economies of
scale, a proxy for the advantage of standardization, and the desire for variety.

Opponents of trade liberalization often fear standards would degrade to the lowest
level. Garbo’s (2002) international convergence in regulation in case of measurable
standards shows that, even without acceptance of reciprocal minimum standards, a
process of iterative adjustment may be triggered by the coexistence of foreign goods with
high standard levels and domestic goods with relatively low standard levels. Focusing on
the case of international heterogeneity of levels of a specific standard, he offers an
intuitive counter-argument to the fear that free trade necessarily implies a "race to the
bottom" of standard levelss. A similar motif of minimum quality standards appears in
Lutz’ (1996) partial-equilibrium model of vertical product differentiation and trade in
which duopolistic firms face quality-dependent costs and compete on quality and price in
two segmented markets. In an unregulated equilibrium, one firm sells high quality
whereas the other sells low quality. Under full harmonization of standards, there always
exists some minimum quality standard that will increase welfare in both regions.
Alternative policy arrangements include national treatment, and mutual recognition.
Under either alternative, standards are found that increase welfare in both regions. Mutual

recognition is always better than no regulation.

 

8 For more on the “race to the bottom”, refer to Wilson, JD, 1996; Levinson, 1996; and Klevorick, 1996.

10

1.2. Trade Agreements

1 .21. Institutional Background

Paragraph 1 of the GATT 1947, Part I, the Article I on the General Most-Favored-
Nation Treatment guarantees that “any advantage, favor, privilege or immunity granted
by any contracting party to any product originating in or destined for any other country
shall be accorded immediately and unconditionally to the like product originating in or
destined for the territories of all other contracting parties,” and in fact prevents formation
of subglobal agreements. However, the WTO members are allowed to form sub-global
agreements under the condition that duties and other trade barriers should be reduced on
substantially all sectors of trade for all members of the agreement, and non-members
should not find trade with members any more restrictive than before the agreement
(WTO 2002)?

Among the advantages of a membership in a trade agreement are enhancement of
terms of trade for members (relative to nonmembers) of an accord as a whole; trade
creation; economies of scale; offset of eventual domestic monopoly power; increased
levels of investments and technology transfers followed by industrialization,
development, and increased rates of growth; increased bargaining power within a region
as well as globally; and the means to consolidate domestic economic reforms and provide
political stability in transition economies (Bhagwati et a1 1998, Schiff and Winters 1997).
The disadvantages include trade diversion; losses in tariff revenue following removal of

trade barriers; necessary investment of resources needed to negotiate and implement an

 

9 Rules for forming clubs among countries - individual political entities - are described by the WTO. On
the other hand, forming agreements among profit maximizing firms resulting in cartels are prohibited in
most countries in the world.

11

agreement; often confusing rules of origin; and additional issues of technical,
administrative, and political nature arising from memberships in several overlapping

treaties (Bhagwati et al 1998, Schiff and Winters 1997).

1.2.2. Literature: Trade Agreements

Research on trade agreements often assumes their exogenous formation. Whether
modeling a new arrangement (i.e., Grossman and Helpman, 1995; Richardson, 1994;
Levy, 1997) or modeling extensions of already existing agreements (i.e., Baldwin, 1993;
Bond and Syropoulos, 1996; Yi, 1996; Andriamananjara, 1999), to our knowledge the
idea of linking agreement formation to the existence of NTBs has not been entirely
pursued.

The concepts of public and club goods provide an elegant framework for analyses
of diverse trade arrangements. Public goods, whose benefits simultaneously affect a
group of individuals, but cannot be divided among individuals, owing to nonrivalry of
benefits and excludability problems, fit the portrayal of global trade agreements (the
WTO). Club goods, characterized by congestion and excludable benefits (Comes et al.,
1996) express the workings of the sub-global agreements. Clubs as voluntary groups are
justified in the literature by pure taste of association (maybe on historical grounds), cost
reduction from scale economies or team production, sharing of public goods, and sharing
of public factors (Comes et a1, 1996). In many definitions of clubs in the literature, club
members share common characteristics, such as income level.

Sub-global agreements exclude the rest of the world; and free movement of goods

and services within such an agreement is closer to a club good in Buchanan’s sense than

12

to a non-excludable public good. Arrangements in globally multilateral organizations
resemble a public good due to the non-excludability of benefits"). Studies using concepts
from public economics in an international setting have been mostly concerned with
security (Olson and Zeckhauser, 1966; Brauer et al., 2000), and lately with the
environment (i.e., Pimbert, 1997; Carraro etal., 1993). A link between agreements and
clubs does occur (i.e., Fratianni and Pattison, 2001; Keohane and Nye, 2001; Kerr, 2002
- none of them agrees on the meaning of a “club”), however, the literature seems largely
to ignore public concepts to explain the trade agreements phenomena.

Fratianni et al (2001) use a simple club theory model to derive the optimum
amount of cooperation while stressing the need for a dominant provider, but their paper
contains mostly lessons from club theory applied to LTGs. Organizations with large and
heterogeneous memberships offer less probability of success in international cooperation
than smaller and homogeneous clubs led by a few players, keeping in mind “inter-block”
differences and rivalries: i.e., EU, NAFTA, and Asia.

Keohane and Nye (2001) discuss a club model of multilateral cooperation. They
use a “club model” as a modus operandi: international organizations operate as clubs of
negotiators, often technically trained, bargaining with one another within specified issue
areas. The club model is very convenient for officials negotiating agreements within issue
areas as it keeps outsiders out: first, officials working on different issue-areas are
excluded from the negotiations; and secondly, the public is confronted with a series of
faits accomplis, making domestic politics easier to manage. Their model can be applied

to negotiators representing heterogeneity and attributes of member countries, although

 

'° In a multilateral agreement -— “big club” — non-exclusion can be treated as a choice, rather than a
necessity.

13

they do not necessarily position themselves in that way. With the “club of negotiators
from the rich countries” still setting the agenda and leading negotiations, other countries
might seek more homogeneous clubs to present their concerns, or — members not
“selected to be in the decision making club are” likely to form their own clubs.

Finally, Kerr (2002) suggests that “since its inception the GATT, and
subsequently the WTO, has been able to operate in a fashion that is more consistent with
a club than an inclusive organization that encouraged the active participation of all its

members”, and its global club model might no longer be appropriate.

1.3. Standards and Trade Integration

1.3.1. Anecdotal Evidence: Standards (NTBs) and Agreements

How do standards and subglobal agreements link? While causality is open to
discussion, anecdotal evidence shows there are a large number of bilateral and regional
agreements aiming at harmonization or mutual recognition of national standards that also
aim at facilitating trade between countries (i.e., Wilson, 1997). Successful agreements
tend to be industry driven: for example, the World Wine Trade Group, consisting of
Argentina, Australia, Canada, Chile, New Zealand, South Africa, and the United States,
signed an agreement on mutual acceptance of oenological practices in 20021 l. A similar
agreement on a bilateral basis is being negotiated between the EU and Australia”. A

further case: in 1995 the governments of Australia and New Zealand signed a Treaty

 

" Full text of The Agreement on Mutual Acceptance of Oenological Practices is available at
http://www.ita.doc.gov/td/ocg/oenological.htrn. Last viewed on July 7, 2004.

'2 From EU’s Bilateral Trade relations. Available online at http://europgeu.int/comm/trade/issues/bilateral
Icountries/australia/index en.htm Last viewed on July 7, 2004

14

establishing a system for the development of joint food standards (Lindenmayer, 1999),
following the Australia New Zealand Closer Economic Relations Trade Agreement
Signed in 198313.

A practical example of harmonization — in terms of eliminating NTBs — is seen in
the current EU enlargement”. One of the first short- and medium-term priorities in each
country was to establish and consolidate standardization and conformity assessment
structures. While the concept of “deeper integration beyond abolition of import tariffs
and quotas, to further measures to remove market segmentation and promote integration”
(V enables, 2000) is by and large not questioned, critical views on potential losses of
national standards banning national identity — i.e., what constitutes “rum”, and how to
address cheeses made from non-pasteurized milk — are seen as well.

The issue of “lost national standards” repeats itself: in 1987 the EU (then
European Community) published its visionary plan on new, standardized Europe of 1992;
a gain of 7 per cent of European income was estimated from harmonization (Emerson
1988, p.6). Even then British were opposed to some parts of it, claiming “brilliant green
mushy peas” and “pink sausages” are part of their national identity, and eventually
succeeded in getting the necessary exemptions (Krugrnan and Obstfeld 2002). Although
the harmonization issue has been on the tables in Brussels at least since the late 19805, it
yet has to be concluded. Over the years a “new” approach replaced the “old” one: instead
of imposing technical solutions, the EU legislation is limited to establishing the essential

requirements which products must meet (EC, 2003).

 

'3 There are three areas of food standards excluded from the operation of the Australia - New Zealand joint
system - the specification of maximum residue limits for agricultural and veterinary chemicals in foods, the
specification of food hygiene practices and export requirements relating to third country trade.

’ Countries joining the EU in May 2004 after the Accession Treaty is ratified are Czech Republic, Cyprus,
Estonia, Hungary, Latvia, Lithuania, Malta, Poland, Slovakia, and Slovenia (hgp://europa.eu.int).

15

The EU recognizes the removal of technical barriers to trade as a necessary
prerequisite of a complete internal market. Rather than establishing definite technical
norms and solutions on a case-by-case approach as it was the instance before 1985, the
legislation now establishes minimal requirements products must meet. Harmonized
European product legislation includes legislation covering conformity assessment bodies,
accreditation bodies, standardization, and market surveillance. Regulations continue to be
voluntary, manufactures therefore remain free to offer, on the Community market,
products meeting other standards or not meeting any, provided that they satisfy the
procedures for assessing conformity laid down by the appropriate Directive (EC, 2003).

The Commission also recognizes potential problems arising from the voluntary
nature of its standards: in its 1998 report, the Commission emphasizes that the voluntary
and independent nature of standards drafting can only be justified if the system is truly
open and transparent and if the standard is supported by all major interested parties and
applied in a uniform way throughout the Community. Consequently, European
standardization is slower in the areas in which the market is not particularly supportive of
standardization, for example in case of goods not suited to be transported, and thus are
only consumed locally. The standardization process in the EU takes place in several
“layers” of national and European standardization bodies mutually cooperating and
informing each other of drafts standards and potential amendments. Where technical
standards are not harmonized, the principle of mutual recognition of national rules

applies (in line with the Cassis de Dijon judgment”).

 

lsThe Cassis de Dijon judgment ensures that “any product lawfully produced and marketed in one Member
State must, in principle, be admitted to the market of any other Member State. For more on the Cassis de

Dijon, please refer to hgpzlleuropaeu.int/comm/intemal market/en/goods/caasiscomm en.pdf. Last viewed
on July 8, 2004.

16

Standards are just one example of domestic regulatory policies. Their
synchronization or at least convergence of positions does not differ from any other policy
integration — or “actions by governments to reduce the market segmenting effect of
differences in national regulatory regimes through coordination, harmonization, or
mutual recognition of national laws, regulations, and enforcement mechanisms”
(Hoekman et a1, 1998). Some of the generic reasons why policies are integrated include,
but are not limited to, lessening hostility; cultural dimension in choosing partners;
political cooperation in spheres other than trade: water basin management, environment,
and infrastructure; economies of scale in negotiating with the world and providing local
public goods. In all cases where policy integration is being considered, the question of the
optimal domain of inter-govemmental cooperation is to be addressed. While in some
instances the optimal domain will be regional (or subglobal — added), in many instances

the optimal domain is global (Hoekman et a1, 1998).

1.3.2. Literature: Standards and Coalitions

Several motifs are repeated in this literature: the most frequent ones are labor and
environment, more recently intellectual property rights (IPR), often from the NAFTA
perspective. For example, Esty (1994) examines the treatment of environmental
considerations in the NAFTA, and concludes that it provides some useful first steps in
efforts to make trade and environmental policies mutually supportive. In the same spirit
Emerson and Nessman (1995) praise the NAFTA for shaping future regional and
international trade negotiations by dealing with both boundary problems and common

standards without dictating uniformity of standards or impinging greatly on each nation’s

17

sovereign control over matters within its own borders. Ederington (2001) shows that in a
repeated game framework, the enforcement of a free trade agreement may require some
convergence in environmental standards across countries when trade is driven by
differences in such standards. He also demonstrates it is more efficient to enforce a trade
agreement by setting tariffs partially to offset differences in policy standards than to
attempt to harmonize standards within environmental side agreements. In the labor arena,
Singh (2002) finds the NAFTA labor side agreement had enjoyed moderate success in
bringing labor concerns to the fore and that predicted negative effects of trade
liberalization — namely the erosion of standards and the harmonization of labor laws to
the lowest common denominator - did not materialize. Lai and Qiu (2003) build a multi-
sectoral North (“developed”) —- South (“developing”) trade model to analyze international
IPR protection, and show both regions can gain from an agreement that requires the
South to harmonize its IPR standards with those of the North, and the North to liberalize
its goods market.

The concept of smaller “alliances” and standards also appears in Gandal and Shy
(2001). Instead of harmonization they focus on a govemment’s incentives to recognize
foreign standards, and allow a game in which countries can form standardization unions.
The outcome depends on the size of network and conversion costs.

Gaisford (2002) examines implications of the biotechnology production and
regulations for the Free Trade Area of the Americas (FT AA), and recognizes that
different countries in the area have different stakes as most of the biotechnology research
was aimed at temperate climate crops. The F TAA may be tempted to take a minimalist

approach, and defer biotechnology regulations to other relevant authorities, such as the

18

SP8 and TBT agreements and the BioSafety Protocol, even though these outside
authorities are, at present, neither fully up to the task posed by biotechnology nor even
mutually consistent. A more ambitious approach, for the F TAA might be, therefore, to
provide interim clarification on important issues, such as that GM and non-GM varieties

need not to be treated as like products, and a short-term precaution.

1.4. Contribution to the Literature

This dissertation fills a gap in the literature in multiple accounts: (1) it focuses on
standards for “like-product” as an illustration of non-tariff trade barriers, and (2) links
sub—global agreements (not necessarily formalized) with countries’ preferences and
positions. We treat standards as representations of consumer preferences only — and leave
out lobbying effects of domestic producers eager to establish their own platform as a
prevalent standard. In all cases we assume launching and enforcement of standards is
costless. Standards in the models are imposed to facilitate consumers’ decision, take
advantage of increased number of varieties, and to a certain extent economies of scale.

Extending the literature on strategic trade policy, Chapter 2 explores standards as
strategic policy instruments by illustrating country’s possibilities to influence terms of
trade by changing the standard at which country offers parts of its endowment for trade.

Unlike other authors, we focus on the issue of like products (effectively
indistinguishable products with the same end use): where usual motifs for standardization
(other than smoothing trade) tend not to apply. However, recent trade disputes and

consequent polarization into sub-global groups — has shown increased interest in process

19

standards, and in like products in general. We do not model standards in a duopolistic
structure of limited number of goods, but extend the usual Krugrnan model of
monopolistic competition and love of variety [in Chapters 3 and 4] to incorporate
standards. Appending a simple, yet well established and commonly understood model
allows us to concentrate on conclusions and implications of the model, rather than
focusing on its mechanics.

The dissertation focuses on standards, and treats trade agreements as a
“byproduct” -— showing that optimal harmonization domain can be sub-global. After
solving the case of standard selection between two countries, we extend the analyses to
the multiple country case, and claim that provision of standards is more likely as a club
good rather than a public good [unlike Casella who explicitly seeks to model standards as
club goods]. We do not seek to model formation of sub-global trade agreements per se:
formation of trade agreements is a natural consequence of lack of consensus on reaching
a global standard, or, showing harmonization is disadvantageous when the differences
between countries are too big. Such a formulation fits nicely with club theory and theory
of social choice, and complements political economy contributions comparing
international organizations with clubs.

Finally, last chapter applies conclusions from the theory essays to trade in

genetically modified products.

20

CHAPTER 2

STANDARDS AS STRATEGIC POLICY INSTRUMENTS

Chapter 1 introduced a specific member from the family of non-tariff trade
barriers: standards as an example of technical barriers to trade. It also suggested motifs
for increased application of non-tariff barriers: with the tariff bound in the WTO
negotiations, countries who wish to manipulate their terms of trade, protect domestic
consumers, etc. are obliged to employ non-tariff measures as a back-door to prohibition
of “traditional” trade barriers. While the practice is certainly discouraged, and often
results in a dispute settlement procedure, current treaties (SPS, TBT, etc.) allow the
application of non-tariff barriers in the name of the protection of human, animal, and
plant health. The Chapter seeks to demonstrate different levels of welfare attainable with
different standards depending on the policy scenario chosen.

In this essay we assume two pure exchange economies with heterogeneous
preferences, and study mechanics of their standard selection under different scenarios
balancing domestic preferences with gains from trade — effectively employing standards
as some sort of a “strategic trade policy” instruments. Namely we consider four different
trade cases”:

1. Case I: choice of standards in a social optimum employing an omnipotent social
planner taking into account the extemality imposed on the other country;

2. Case 11: choice of standards in a “myopic” equilibrium without terms of trade

effects;

 

'6 Due to CD structure of the preferences, autarky in which each country would consume own good only is
not an equilibrium: regardless at what standard the domestic good would be offered.

21

3. Case 111: choice of standards in a Nash equilibrium game in standards setting with
terms of trade manipulation; and
4. Case IV: choice of standards in a Nash bargaining solution.

Trade literature defines strategic trade policy as a trade policy that conditions or
alters a strategic relationship between firms, implying that strategic trade policy focuses
primarily on trade policy in the presence of oligopoly. The key point is that strategic
relationships between firms introduce additional motives for trade policy, over and above
terms of trade and other effects that arise in all market structures (Brander in Grossman
and Rogoff, 1995). Strategic trade policy also takes account of imperfect competition and
increasing returns (Krugrnan and Smith, 1994). While standards selection per se does not
consider anything fiom above specifically, it provides governments with a tool to balance
two effects: consumption and terms of trade effects, and thus introduces strategic reasons
into standards selection between countries (not firms). The choice of standards -
effectively a trade policy — shifts the shares of goods consumed at the domestic and
foreign markets — and results in different gains in the form of terms of trade effects.

The model corresponds to decisions countries are facing when trading like
products (products with the same tariff classification and end use), or considering

different production processes which impact countries’ willingness to pay for goods.

2.1. Model Description

The chapter abstracts from production by assuming exogenously-given

endowments of goods. Two goods are available in the world, and each country is

22

endowed with one good only: Home is endowed with X units of good X; Foreign is
endowed with Y units of a good Y I 7 . We will explore cases of identical and different
endowments across countries. Endowments are fixed: the choice variable is the quality at
which countries offer their endowments. The origin of the good (Home or Foreign)
indicates two individual goods, rather than denoting geographical attributes of goods that
would be otherwise substitutes to a certain degree (for example, wine from France and
Italy). Trade is guaranteed by assuming a Cobb-Douglass structure of preferences in each
country: utility in autarky with only one good consumed is zero. We consider a
representative consumer problem in each country”. After a monotonic transformation,

(the natural logarithm of) Home’s utility function is given by:

H
(2.1A) U =aHlan+flHlnyF

while the Foreign utility is given by

(2.18) UF =aFlan+flFlnyF
where x and y stand for consumption of the Home and Foreign good, respectively. The
subscript denotes the consumption point (Home or Foreign). Define exponents (or

preference weights) as:

1

(2.2A) aH h =————
() 1+(h—H)2

l

(2.2B) a1: h =————
() 1+(h—F)2

 

'7 In text upper case letters (X and 1’) denote supply; lower case letters denote demands.
" Size considerations are addressed via endowment size.

23

1

(23A) ,5 f=——"
H( ) l+(f-H)2

1
(2.33) flF(f)=-————2-
1 + (f - F)
where H and F stand for exogenous stable Home and Foreign quality preference
parameters; h and f are endogenous Home and Foreign quality choice parameters

(standards)19. Holding relative prices, x,- and y,- constant, and assuming x,- > I , y,- > I the

. . . . . . . 6 - 6 - .
utrlrty in each country rs moreasrng 1n (1,, [3,: B—U—l > 0 and—E'— > 0. Srnce x,- and y, are
“i i

chosen to maximize utility, allowing consumers to re-optimize after the change in a; and
6,- can only result in higher utility. Note that a,- terrns (Equations 2.2A and B) depend only
on the home’s choice parameter h, while the [3, terms (II.3A and B) depend only on f
Quality parameters - call them standards - can be associated with quality attributes, food
safety standards, etc. - that is, any attribute from which a diversion can be assessed.20 For
simplicity assume F and H are borderline points of an interval along a number line, with
F being less than H (potential interpretation includes tolerance levels in the system).
“Standards” are measured in terms of the Euclidian distance from the countries’
respective borderline preference parameters: F for Foreign, and H for Home. We only
consider cases where F and H are different: the case when F and H are identical is trivial,
and countries would pick a standard identical with the common tolerance level. Setting a
standard equal to the preference parameter (Home at H, Foreign at F) guarantees that

weights (or and ,8) equal one. Moving away from domestic preferred “quality” parameter

 

'9 In text we will refer to F and H as preference parameters, and f and h as choice parameters.

2° Presented model is not applicable to horizontally differentiated goods (none of the attributes is superior).
While in the model we abstract from considering the potential superiority, it is true that in order for the
model to work, it has to be possible to “measure” the difference between desired and offered standards.

24

causes the weight assigned to a consumption of a particular good to decrease. Figure 2.1
illustrates the behavior of a” and a; as a function of h: a; reaches its maximum at h = F,
a” at h = H. Both curves intersect at h = (F +H)/2. Figure 2.2 provides the same graph for

£5 as a function of foreign choice parameter f

 

 

 

 

 

 

 

 

 

1.0
1
1+1?2
1
1+H2 .
F+H H h
F 2

Figure 2.1: Behavior of a Exponents as a Function of h.

 

 

 

 

 

 

 

 

1.0
1 t
1+F2 flHU)
1 j flF(f)
1+H2 J
F F:H H f

Figure 2.2: Behavior of [3 Exponents as a Function off

25

Choosing appropriate h and f constitutes a “strategic policy tool” in the case
presented: by choosing h (at Home) and f (at Foreign) countries balances between
domestic and foreign demands - and thus influence their “terms of trade” (TOT). We
assume each country is obligated to choose one “standard” only (h or f) at which it will
supply its endowments - and rule out cases of endowment segregation to tailor part of the

endowment to suit taste in both countries. Substituting for weights in Equations 2.2A and

 

Bweobtain:

(2.4A) UH =_1__21an+ 1 21nyH
1+(h-H) 1+(f—H)

(2.43) UF =_1__21an+-———1—-2—InyF
1+(h—F) 1+(f-F)

Normalize the price of the Foreign good to be the nurneraire. Thus, representative
consumers face a budget constraint of:

(2.5A) px H + y H = E H in the Home country

(2.5B) px F + y F = E F in the Foreign country

where p is the price of the domestic good, and E3 and Er are values of their respective

endowments:
(2.6A) E H = pX
(2.6B) E F = Y

By Cobb-Douglass preferences

(2.7A) pxi = giEi

(2.73) yi = (1’61')Ei

26

where

“i

a; +fli

 

(2.8) 6i =

is the expenditure share for x, i = H or F. Substituting for demands into Equations 2.4A
and B, and simplifying the expression, the indirect utility function becomes:

(2.9) V,- (p,E,') = ai 1n6i +fli ln(l—6l,-)-a,- lnp+(a,- +,B,°)1nE,-

for i = H, F. Substitute values of endowments from Equations II.7A and B into [1.8:
(2.10A) VH (p,X) = my ln6H +,BH ln(1—6H)+,BH lnp+(aH +,6H)lnX

(2.10B) VF (p,Y) = a}: 111017 +31: ln(1—0F)-ap lnp+(aF +flF)InY
Notice the Home’s indirect utility (2.10A) is increasing in price, while the Foreign

indirect utility (2.10B) is decreasing in price.

2.2. Case 1: Social Planner

We commence with the social planner problem to account for “extemalities”
countries are imposing on each other by their selection of standards as an efficient
benchmark case. Due to the structure of the preferences, to achieve positive utility,
consumers in each country must consume some definite amounts of both goods
regardless at what standards they are offered. The extemalities take the form of
consumption that is lower than it would be if a different set of standards was chosen. The
social planner’s selection chooses a set of standards that minimizes extemalities countries

would otherwise impose on each other.

27

The social welfare function consists of a sum of individual countries utilities
across countries, assuming each country consists of a number of consumers (X consumers

at Home, 1’ consumers in the Foreign country), each endowed with one unit of good”:
(2.11) U=X(aHlan+flHlnyH)+Y(aFlan+,BFInyF)

Goods’ markets clear:

(2.12A) XxH+YxF =X

(2.123) XyH+YyF =Y

Consequently the demand functions are:

 

 

 

 

(2.13A) xH= XO‘H
XaH+YaF
(2.13B) xF= X‘ZF
XaH+YaF
(2.14A) yH= ”311
XflH+YflF
(2.14B) yF= Y’BF
XflH+YflF

After substituting for demand functions from Equations 2.13A — 143 into 2.11, the social

indirect utility function is:

_ XaH YflH
W —X{aH IanaH +YaFf+flHlnIXflH +YflF ]}+

XaF YflF
Y{aFlnI:XaH +Yap]+fiFlnl:XflH +Yflp]}

 

 

(2.15)

 

 

 

2‘ Consideration of a non-weighted social welfare function of two representative agents and addressing the
size considerations through the size of the endowment of each agent results in the social planner splitting
the difference between the standards (that is, choosing both standards in the middle of the interval), and
inevitably leads to large transfers from the larger to smaller country.

28

The social planner maximizes social welfare by choosing the choice parameters
for both countries, while taking into account the extemality the selection imposes on the
other country. Thus, the FOC (simplified by incorporating definitions of exponents from

2.2A - 3B) are:

 

 

(2.16A)§Z=2 a}2{(H—h)Xln XO‘H +a12:(F—h)Y1n X017
6h XaH-l-Yap XaH+YaF

 

 

?K_ 2 _ YflH 2 _ YflF
(2.16B) af —2{,BH(H f)Xln[X,BH+Y,6F:l+flF(F f)Ym[XflH+YflF]}

Notice that the social planer considers marginal impact on “one set” of exponents
at the time: i.e., he weights the impact of the choice of h on as independently from 35.
Appendix A.1 lists the second order conditions for the problem when countries are
endowed equally.22 Whether the objective function is convex or concave on the interval
bordered by Foreign and Home preference parameters depends on the Euclidian distance

between them.23 Thus, we prove two propositions depending on the respective distance.

Proposition 2.1: When endowments are the same across countries, and the difference
between Home and Foreign preference parameters (F and H, respectively) is sufficiently

large, the social planner splits the difference between Home and Foreign preference

parameters, settingf=h =(F+H)/2.

Proof: When the endowments across countries are identical, the first order conditions

(Equations 2.16A and B) reduce to:

 

22 We briefly touch upon the social planner solution when endowments are different in Appendix A.2 by
providing numerical examples and implying some conclusions.
We ignore the case when the objective function is neither convex nor concave.

29

(2.1M) Ez=zx{a%{(H-h)ln[——9—H——]+a%(F—h)ln[_gf_]}

6h aH+aF aH+aF
6W 2 [3H 2 flF
2.17B —=2X H— ln —— F— In ——
( ) 6f {flH( f) [flH‘l' F]+flF( f) [16H+:6F]}

By assumption(H — h) > 0 , and (F — h) < 0. Therefore, for the first order condition with

respect to Home choice parameter (h) to hold (Equation 2.17A), it has to be true that:

(2.18) a%1|(H - h)| 1n [—aHafaF ] = a}; |(F — h)l 1n [.aHafaF]

When h = F 3 H => IH - h| = |F - h| . Consequently from Equation 2.2A and B,

 

2 2

05H _ “F , and “H = “F' A similar line ofreasoning

a =a .Thus, ..
H F aH-l-ap aH+aF

holds for ,8” and fl;: ,8 H = ,6 F when f = (F + H ) / 2 . Second order conditions are in the

Appendix A. 1.

Proposition 2.2: When endowments are the same across countries, and the difference
between Home and Foreign preference parameters (F and H, respectively) is sufficiently
small, the social planner choice of social welfare maximizing standards (choice

parameters) coincides with countries’ preference parameters: that is, h = H and f = F.

Proof: The simplified first order conditions for the welfare maximization problem
(Equations 2.17A and B) apply. However, based on the exposition of the second order
conditions in the Appendix A] when the difference between the Home and Foreign

preference parameters (F and H) is “small”, the critical point in the middle of the interval

30

is welfare minimizing, as the social welfare function is convex on the interval. Other

critical points lie outside the (F, H) interval, and thus are prohibitive based on economic

intuition. Therefore, countries’ preference parameters are identical with their choice

parameters.

2.3. Case 11: Myopic Equilibrium

By myopic (antonym to a perfect foresight) equilibrium we mean a situation
where each country is maximizing its indirect utility function without considering the

impact of a change in the standard on the terms of trade - or, it assumes price is constant.

Proposition 2.3: In the myopic equilibrium the social welfare maximizing standard is

identical with the country’s preference parameter (at Home: h = H, in Foreign: f = F)

Proof: Addressing choice of the Home country first we take the first derivative with
respect to h (the domestic choice parameter) of the Home’s indirect utility function (Eq.
2.9A), and simplify the notation using Equations 2.2A — 8 to obtain:

(2.19) %—=-2(h—H)agz (11.9,, +lnX)

Setting the first order condition equal to zero (keeping in mind a H , 0H are functions of

h), we solve for the following stationary point:

 

(2.20) h=H— —2—(f-H)2+X+(f-H)2X

31

 

(2.21) h=H+\/—2-(f-H)2+X+(f-H)2X

(2.22) h = H

Appendix A.3 evaluates the relevant second order conditions, concluding that Home’s
indirect utility function without terms of trade effects reaches its maximum at h = H, and
minima at the other two stationary points. Figure 2.3A schematically graphs the behavior

of Home’s indirect utility function taking price and F oreign’s choice of standard as given.

A
VH

 

 

Fi éH h

Figure 2.3A: Home: Indirect Utility Function without Terms of Trade Effects.

In the Foreign country the FOC (differentiating Equation 2.103) becomes:

dV
(2.23) 7;"- : -2(f—F),B% (1n[1-9F]+1n Y)

Solving for stationary points, we obtain:

 

2

(2.24) f=F- —2—(h—F) +Y+(h—F)2Y

32

 

(2.25) f=F+\/—2—(h—F)2+Y+(h—F)2Y

(2.26) f = F

As in the case of Home country, the Foreign indirect utility function reaches a maximum
at f = F, and minima at the other two stationary points (SOCs are evaluated in the
Appendix A.3). Figure 2.33 illustrates the behavior of foreign indirect utility function

without terms of trade effects (without considering the effect on price).

VF

 

 

 

- - - -
-—I-ON— . ...-_—._...~_...—......_....... .. .. -

F

Figure 2.33: Foreign: Indirect Utility Function without Terms of Trade Effects.

Thus, in the myopic equilibrium - defined as countries ignoring their potential

impact on price, and forsaking terms of trade effects, countries identify their choice

parameters with their preference parameters. QED.

33

2.4. Equilibritun Price

By Walras’ Law, we only need to look at one market. For the Home-good (X)
market to clear, the sum of individual country demands has to equal world supply:
(2.27) xH+xF=X

Substitute for individual country demands to obtain:
(11.28) 91115,, +9-E-EF = X
P P
From Eq. 2.6A and B substitute values of endowments for incomes:
9F
(11.29) 6HX +— Y = X
P

and finally derive an expression for equilibrium price defined as a price of the Home
good (with a safety parameter h) in terms of a numeraire — that is the Foreign good (with
a safety parameter f).:

(2.30) p =—0-E—-’i
l— 6H X

Price is a function of expenditure shares (63) multiplied by relative endowments. Recall

the 6, represent expenditure shares (Eq. 2.8), and are defined as:

 

aF (h)
2.31A e =
( ) F a1r(h)+i61r(f)
(2.31B) “H (h)

 

6H =
air 0?) + flH (f)
Starting at h = H, a small change in h has a second order effect on a” (at h = H,

flu—H— = 0 ): m < 0. So, a small reduction in h causes a; (and therefore 6;) to increase

dh dh

34

but does not change 6”. Therefore, price (p) increases. The situation is reversed at h = F.
Here, a small increase in h does not affect a; (and therefore 6;), but does increase a” (and
therefore 6”). With higher 6H, price increases. Appendix A.4 evaluates the F OCs if the
price at each border point. Since the price of the foreign good serves as a numeraire,
terms of trade from the Home perspective improve as the price of its good increases
(thus, p increases). F oreign’s terms of trade improve as p decreases (I/p increases).
Figure 2.4A graphs price (Equation 2.30) as a function of Home’s choice
parameter (h) given the Foreign standard (f). Similarly, Figure 2.4B graphs price as a
function of f given h. The precise location of stationary points and intersection points
depend on the other country’s choice of standard (for h), as well as relative endowments

across countries.

 

 

 

3“!

F H

Figure 2.4A: Price as a Function of h.

35

 

 

 

\V

;F H

Figure 2.43: Price as a Function off.

Substituting for price (from Equation 2.30) in the indirect utility functions

(Equations 2.10A and B), we obtain indirect utility functions with terms of trade effects:

(2.32A) VHt(f,h)=aHln6H+flHln9F+aHlnX+flHInY
(2.328) VF! (X,Y)=aF ln(1-6H)+flp ln(l—9F)+ap lnX+flF lnY

where the subscript t indicates incorporation of terms of trade effects.

2.5. Case Ill: Nash Game

In a Nash game we consider the Home and Foreign countries to be players
choosing h and 1, respectively (choice parameters as strategies) from a continuum of
potential strategies to maximize their payoff function, consisting of their indirect utility

function with terms of trade effects, in a one shot simultaneous game given the strategy

36

of the other country. Nash cooperation strategy requires moving away from “free trade” —
that is without terms of trade affects. Such a move decreases the indirect utility without
terms of trade effects, while on the other hand improves the terms of trade.

We solve the first order conditions simultaneously. Instead of stationary points,
we focus out attention on demonstrating that (1) potential equilibrium candidates lie
between F and H, border points excluded; (2) if endowments across countries are the
same, countries’ choice parameters will lay in an identical distance from their preference
parameters; and (3) if the endowments differ, countries’ choice parameters position
relative to their preference parameters will change to reflect the difference in
endowments. The Appendix A.5 (Tables A.4 and 5) shows numeric examples sharpening
the intuition gained from the models. We start with proving Nash Equilibrium is on the

(F, H) interval.

Proposition 2.4: Regardless the countries’ sizes, the candidates for Nash Equilibrium in

both countries lie on the open interval (F, H) .

Proof: Note that in a general form the indirect utility (Equation 2.10A) is a function of:

 

(2.33) VH = VH (h, f,p, EH)

where EH was defined in 2.6A as E H = pX = px H + y H . Price (2.30) itself is a function
of endogenous choice parameters and exogenous endowments (which can be left out)
(2.34) p = p(h,f)

In a stationary point a total derivative of the objective function (indirect utility with the

“TOT effect”) is zero:

37

(2.35) 11551 = 0
dh

The total derivative equals a summation of partial derivatives:

5VH +5VH5_p+5VH 5EHfl7__

(2.36)
6h 6p 6h 6EH 6p 5h

 

Rearrange the terms to obtain:

 

 

(2.37) 9111+ 5V” + 5V” X 5—p = 0
6h 6 p 6E H 6h

Apply Roy’s identity:

(2.33) éfii = -151/1 x
5p 5EH

and substitute to 2.37:

(2.39) 5i, flXiKH, 23:0
6h 6E H 6E H 6h

Finally rearrange to:

(2.40) 5V” +:VH ”(X- x1112) =0
5h 6h

Preference points F and H divide the number line into several subintervals. The indirect
utility function is always non-decreasing in income. Earlier we showed that keeping
everything else constant, the first expression in the Equation 2.40 (indirect utility function
without terms of trade effects, Figure 2.3A) is maximized at h = H, and price (Figure
2.4A) is increasing at h = F, and decreasing at h = H. Table 2.1 evaluates Equation 2.40

on each potential subinterval.

38

Table 2.1: Location of a Nash Equilibrium Candidate: Home Country

 

 

 

 

 

 

HOME %’— +g—Z’; (X-xH) $1; =0
he(0,F) + + + + NO
h=F + + + + NO
he(F,H) + + + 2 YES
h=H 0 + + - NO
he(H,oo) - + + - NO

 

 

 

 

 

 

 

 

Based on Table 2.1, it is clear the only potential subinterval where the indirect

utility with terms of trade effects can reach its stationary points is on the open
interval(F,H).
Following the same framework for the Foreign country (keeping in mind E; = X p)

we derive:

5V}: _ 6V}:
5f 5XF

 

(2.41) (xp)%=0

The Foreign indirect utility function without terms of trade effects (Figure 2.3B)
is maximized at f = F. The Foreign country seeks to minimize the price (recall the
Foreign good serves as a numeraire): thus price (Figure 2.4B) is decreasing at f = F, and
increasing at f = H. The results are summarized in Table 2.2.

As in the Home country, in the Foreign country we showed that any equilibrium

candidates will be from an open (F, H) interval. QED.

39

Table 2.2: Location of a Nash Equilibrium Candidate: Foreign Country

 

 

 

 

 

 

FOREIGN 5%;— - 53?; as g; = 0

f e (0,F) + - + + - NO
f = F 0 - + + - NO
f e (F,H) - - + + ? YES
f = H - - + , + + NO
f e (H,oo) - - + + - NO

 

 

 

 

 

 

 

 

 

2.4.1. Nash Equilibrium with Identical Endowments

In the world of identical endowments, we conjecture and prove the Nash
I Equilibrium is symmetric (that is, equal distances from the border points). Numeric

simulations are provided in Table A.4, Appendix 11.5.

Propojsmon 2.5: Returning to a specific functional form with terms of trade effects
(Equations 2.32A and B), when endowments across countries are identical, the Nash
equilibrium standards (that is, standards chosen by the government) will be in equal

Euclidian distances from the border points (and “myopic” equilibrium points) F and H.

Proof: The F OCs reduce to:

dVHt

2.42A ——
( ) dh

=2{(1—6F)aFflH(F-h)+a121(H-h)[(l—6H)+ln6H+lnXJ)

40

dVFt

(2.42m?— = 2(0),"):pr(H—f)+p12r (F-f)[6F +1n(l—6F)+lnY]}

Notice parts of each FOC are identical with the first derivative of the indirect
utility function without terms of trade effects (Equations 2.11 and 15). We showed earlier
the indirect utility function without the terms of trade effect is maximized in their
respective border points: Home’s indirect utility in h = H, and Foreign indirect utility inf
= F (Figures 2.3A and B). Additional parts of the derivative (Equations 2.42A and B)
come from inserting the price to the indirect utility function. We have shown earlier in
the general setting that the stationary point optimizing the indirect utility function with
the terms of trade effects must lie between the border points of F and H, and will
“balance” the marginal loss in welfare with the marginal gain in terms of trade for both
Home and Foreign countries.

Due to the nonlinearity of the first order conditions, (h and f entering in more than
one argument, preventing the use of inverse functions), standard algebraic solution
techniques do not lead to a closed-form solution for the stationary point.

The larger the countries’ endowments get in the absolute terms, the closer the

choice parameters (f h) approach the preference parameters (F, H). The solution remains
. . dV . . .
1nter10r. Suppose for example that X —) oo . Then a = 0 => (H - h) -) 0 . Srmrlarly, wrth

the increasing difference between the preference parameters, countries’ choice
parameters approach their preference parameters. When (H - F) is large, the distance
between F and h increases as well. To balance the first order condition when the effect of

preference and choice parameters on the exponents of the utility is balanced by identical

endowments, (H - h) —> 0.

41

For an analytic proof define:
(2.43A) IF- fl = |H -—h| = |m|
(2.433) |F -h| = |H — f] = |n|
Due to the structure of exponents (Equations 2.2A - 3B), it is true that:
(2.44A) aH = ,6}: = c
(2.448) at}: =flH =d
Identical exponents translate into the Home’s expenditure share on home good being

equivalent to Foreign’s expenditure share on the foreign good, and vice versa:

(2.45A) 0H: “H = flF =1-6F=e
aH+flH aF+fiF

 

 

011? HH
(2.453) (91.— = = =1_9H = g
0F +617 aH +flH

Substitute shorthand labels from Equations 2.43A - 45B into Equations 2.42A and B:

(2.46A) 1;? = 2{ed2 |n|+c2 |m|[g+1ne+1nx]}
dVFt 2 2
(2.463) 7:2(ed |n|+c |m|[g+lne+lnY]}

Notice Equations 2.46A and B are identical except the size of the endowment. We
use proof by contradiction. Assume countries with different endowments would choose
their standards to be in the same distance from the domestic preference parameters.
However, if indeed the equilibrium points are positioned equivalent distances from the
border points, then it has to be true that the Home and Foreign endowments are identical
in order for the FOC in both countries to be zero, and welfare — consisting of indirect

utility function adjusted for terms of trade effects — is maximized. QED.

42

2.4.2. Nash Equilibrium with Different Endowments

Now we deliberately keep the endowments unequal. Table A.5 in the Appendix
A.5 shows numerical simulations. They suggest the country with larger endowment has
more gravity, and its choice parameter gravitates towards its preference parameter. That
is, country with the smaller endowment is choosing its standard to be positioned a larger
Euclidian distance from its preference parameter than the country with larger endowment

is doing relative to own preference parameter. Therefore,

(2.47A) X<Y=>|H—h|>|F—f|
(2.473) X>Y=>|H—h|<|F—f|

Graphically both situations are illustrated on Figure 2.5.

 

 

F

l f

L V J l J
X

 

 

 

[F f h Hl

L__.,_J L V J

X<Y=>|H—h|>|F-f|

 

 

 

 

 

Figure 2.5: Nash Equilibrium with Country-variant endowments. (not drawn to scale).

Proposition 2.6: When endowments differ across countries, in a Nash game the Euclidian
distance between choice and preference parameters of respective countries differ, with

the more well-endowed country positioning its choice parameter closer to its preference

43

parameter relative to the other country (as outlined in Equation 2.47A and B, and Figure

2.5).

Proof: Based on Tables 2.1 and 2.2, despite the countries’ relative size, the Nash

Equilibrium occurs in an open interval (F, H). None of the FOC — Home or Foreign —

are maximized in the border point of the interval, and both countries gain as the they
move away from their reference points. Transforming unequal endowments into the price
equation (2.30) consisting of a ratio of preference weights multiplied by the ratio of
endowments, differently sized endowments imply the price changes directly with the
ratio of endowments. Assume Home endowment exceeds the Foreign endowment N
times:

(2.48) X=NY, N>l

Consequently, taking the natural logarithm of both sides,

(2.49) lnX=ln(NY)=lnN+lnY

For a more instructional proof, we use FOCS which have not been treated by substituting
for parameters based on Equations 2.2A — 3B and 2.8, allowing a more detailed
exploration of changes in f and h. After substitution from Equation 2.49, the “untreated”

FOCs are:

44

(2.50A)

dVH, _ (1n(1+(H—f)2)-1n(2+(H-f)2+(H—h)2)+1nr+1nN)(H_h)

dh (1+(H_f)2)2

F—h H—h
+

(1+(H—f)2)(2+(F—f)2 +(F—h)2) (1+(H—h)2)(2+(H—f)2 +(H—h)7-)

(2.50B)

dVFr _(rn(1+(F-h)2)—1n(2+(F—f)2 +(F—h)2)+1nr)(F—f)

df (l+(F-f)2)2
F—f + H-f
(l+(F—f)2)(2+(F—f)2 +(F—h)2) (1+(F-h)2)(2+(H-f)2 +(H-h)2)

 

 

By construction (H > F, and f ,h e (F ,H )and referring to Figure 2.5, it follows that:

(2.51A) F—f=A<0
(2.513) H—h=a>0

F— H-h

(2.52) | f|>| |
lAl>|a|

Similarly, for the relationship across own preference and “foreign” choice parameter:

(2.53A) F-h= 3 <0
(2.533) H—f =b > 0
F—h H—

(2.54) l M fl
|B|>lb|

For now ignore that IA] > |a| and|B| > lb]. Rewrite the FOC for Home (2.50A) such that:

45

A(1n(1+32)-1n(2+A2+32)+rnr)

 

 

 

 

AlnN +
2 2 2 2
(2.55A) (1+A ) (1+A )
A + b =0
(1+A2)(2+A2+Bz) (1+Bz)(2+a2+b2)
And for the Foreign:
a(rn(1+b2)—rn(2+a2+b2)+1nr) a
2 +
1+02 (1+a2)(2+a2+b2)
(2.558)
B
+ =0
(1+b2)(2+A2+BZ)

Investigating the FOCS above (2.50A and B), we recognize that if |a| = M

and [bl = |B , as it was the case earlier, the Home’s FOC (2.55A) contains an additional

 

term enclosing the relative difference in the size of endowments between the Home and
Foreign countries. Thus, even if the FOC in (2.55B) for Foreign was satisfied, the
Home’s FOC contains an additional positive term, and therefore cannot be satisfied.
Consequently, if both first order conditions are to hold simultaneously, the equality
between a and A does not hold, and indeed the countries will choose their choice

parameters in an unlike distance from their preference parameters. Now we show that

[a] < [Al and consequently [bl < IBI. Rewrite the inequalities such that:

(2.56A) A = a + Aa

(2.56B) 3 = b + Ab

46

where A < 0, B < 0. Consider a and b to be actual differences between the points on the
number line, not absolute values or Euclidian distances. Rewrite the FOCS (2.55A and B)

to incorporate the substitution:

(2.57A)
(a+Aa)lnN +(a+Aa)(ln(l+(b+Ab)2)-ln(2+(a+Aa)2+(b+Ab)2)+1nY)
(l+(a+Aa)2)2 (l+(a+Aa)2)2

(a+Aa) + b :0

(1+(a+Aa)2)(2+(a+Aa)2+(b+Ab)2) (1+(34Ab)2)(2+a2+b2)

 

 

 

 

+

a(1n(1+b2)-rn(2+a2+b2)+rnr)
+

(1+a2)2 (1+a2)(2+a2+b2)
(b+Ab)
(1+b2)(2+(a+Aa)2 +(b+Ab)2)

a

 

 

(2.573)

+ =0

 

Rearrange the terms in the FOC for Foreign country (2.52B) such that:

a ln(1+b2)-ln(2+a2+b2)+lnY l
(2+az+b2)

(2.58) (1+2?) (H02)
(b+Ab)

+(1+b2)(2+(a+Aa)2 +(b+Ab)2) = 0

+

 

 

Recall by definition a > 0. If Y > 2 (as we already imposed earlier), then

47

1n (1 + b2 ) — ln (2 + a2 + b2 ) + 1n Y is positive. To show it, consider a borderline case

when a2 and b2 attain their maximum values (representing the entire (F, H) interval).
The expression then becomesln(l +(F—H)2)—ln(2+(F-—H)2 +(F—H)2)+lnY,
which reduced to —In 2 + 1n Y . If Y > 2, the expression above positive. If lal and lb] do not

stretch over the entire (F, H) interval, the expression is greater than earlier, and is

positive as well. Further analyzing the foreign F DC in terms of signs:

(2.59) {+}[+]+ 0’ + Ab) ._. 0

(1+b2)(2+(a+Aa)2+(b+Ab)2)

 

(b+Ab)

2 2 must be negative. However,
(1+b2)(2+(a+Aa) +(b+Ab) J

 

For the equality to hold,

by definition, b > 0. Therefore, Ab must be negative, and in fact in absolute value

exceeding the absolute value of b. Recall that F — h = B < 0 and H - f = b > 0. If

B = b + Ab , then Ab < 0. Consequently, if lbl < lBl, then by construction lal < [A], and
countries choose their respective choice parameters in different Euclidian distances from

their preference parameters. QED.

2.6. Case IV: Nash Bargaining Solution

Unlike the Nash Equilibrium, the Nash bargaining solution satisfies additional set
of conditions - Pareto efficiency, symmetry, invariance, and independence of irrelevant

alternatives (i.e., Mas-Colell, Whinston, and Green, 1995). In the model presented, the

48

Nash Bargaining Solution without weights is a bilateral bargaining problem between two
countries, Home and Foreign, choosing their choice parameters (h and f, respectively), to

maximize their Nash product (NP):
(2.60) NP=(VH(f,h)-A)(VF(f,h)—B)

where VH([ h) and VF (f h) are identical with social welfares (equivalent to an indirect
utility function with price effects, Equation 2.32A and B) of the Home and Foreign
countries in the Nash Bargaining Solution. A is a Home country’s threat point; B is a
Foreign country’s threat point. Under certain conditions the threat point would be autarky
utility - however, due to the structure of the utility function considered in the model, we

consider the following threat points:

(2.61A) A = VH(h = H)
(2.613) B=VF(f=F)

That is, in the threat point each country sets its choice standard to be equivalent to its
preference standard regardless other country’s actions.

Due to complexity of the first and second order conditions we do not address the
problem analytically, and present the problem based on the numerical simulations listed
in the Appendix A.6, Tables A6 and 7.

The simulations suggest that, like in the social planner problem, when the
differences between countries’ preference parameters are small and/or the endowments
are small, the Nash product is maximized in the respective preference points (H and F).
In all other cases the Nash bargaining solution is achieved in the identical Euclidian
distance from the respective preference parameters. However, the equilibrium points

approach the center of the interval - the outcome of the social planner problem — more

49

closely than in the regular Nash equilibrium. Often we observed “overshooting”: that is,

while keeping the Euclidian distance between countries’ preference and choice

parameters identical (IH — h] = [F — f I ), this Euclidian distance exceeded the midpoint

between the preference parameters. In case of differing endowments, the outcomes of the
simulations differ depending on the gap between preference parameters as well as the
ratio of the endowments. When the gap between the parameters is sufficiently large,
distance between the choice parameter of the smaller country from its preference
parameter is larger than a similar distance for the larger country (schematically graphed

on Figure 2.6). Nevertheless, both choice parameters approach the midpoint.

F
l f f HI
L L4 4

 

 

X>Y=>|H~h|>|F—f|

 

 

 

Figure 2.6: Nash Bargaining Solution, Country-variant Endowments (not drawn to scale).

2.7. Chapter Conclusions

This chapter explored application of standards as a strategic trade policy
instrument between countries endowed with different goods wishing to trade. It
investigated four different cases: (1) social planner choosing a Pareto optimal allocation
recognizing the extemality countries impose on each other by their standard selection; (2)

countries seeking to maximize their welfare without exploiting terms of trade effects in a

50

myopic equilibrium; (3) countries seeking to maximize their welfare in a Nash type game
recognizing the strategic interaction and power they exercise on terms of trade effect, and
finally (4) countries maximizing Nash product without weights in a complete Nash
bargaining solution.

In the first case — an omnipotent social planner maximizing welfare — the welfare
maximizing point depends on the difference between countries’ preference parameters.
When the difference between countries’ parameters is small, social welfare is maximized
in the border points. However, when the difference between countries’ parameters is
large, the social planner sets standards that split the difference. In the myopic equilibrium
countries choose their choice parameters to be identical with their exogenous preference
parameters. The outcome of the third case - welfare maximization with terms of trade
effects using Nash equilibrium — depends on the relative endowments of the countries.
When the endowments are identical across countries, countries choose their standards
(choice parameters) in the same Euclidian distance from their preference parameters
(border points). When the endowments differ, the country with the abundant good
exercises more power in selecting the standard, and places its standard closer to its
reference point of its preference parameter relative to the location of the country with
scarce good. Finally, in the Nash bargaining solution, depending on the gap between
preference parameters, and the ratio of endowments, in general the choice parameters

approach the midpoint between the preference parameters.

51

CHAPTER 3

DIFFERING STANDARDS, WELFARE, AND SUB-GLOBAL AGREEMENTS

The chapter explores endogenous standards-setting in a Krugman-like model of
monopolistic competition with constant elasticity. Consumers desire more varieties to
less, but also discount consumption of “qualitatively” different brands (“love of variety”
combined with “ideal quality” approach). We consider two questions: (1) potentially
welfare worsening consequences of (unilateral) harmonization and (2) differences in
standards necessitating formation of small trade agreements as a second best option when
common standard is required to conduct trade.

The model is consumer-focused: consumers in autarky have access to the same
ntunber of equally priced varieties regardless which attribute they prefer. We explore the
nuances of standard setting in a two-country world, and demonstrate non-standardized
trade always improves welfare, while standardization to only one quality might be
inferior to autarky. Imposed standardization in a multiple country setting may result in
countries searching for alternative ways to secure (at least some) gains from trade and

forming agreements with partners with similar preferences.

3.1. General Model Setup

We build a general equilibrium model of country-variant standards based on a
conventional monopolistic competition framework (i.e., Dixit and Stiglitz, 1977 or

Krugrnan 1979, 1980). Products are differentiated not only in the variety space, but also

52

in the standards space“. For simplicity assume “discreteness” of the standard: i.e., 0 or 1:
the product either does or does not have a certain attribute. Each variety can be produced
in two specifications, i.e. with one of the two mutually exclusive characteristics. For
example, fruit can have the attribute “organic” or it can have the attribute “blemish-free”,
with “organic” and “blemish-free” being mutually exclusive consequences of the
production process used (organic or conventional fruit production). Standards are
modeled as a vertical distinction within a variety; a “higher” standard represents a higher
quality in the domestic quality ranking, although with different utility fimctions foreign
consumers will have different relative valuations of the high- and low-quality goods (e. g.
Gabszevicz et al., 1981)”. From an individual country perspective a preferred standard
implies an enhanced product, and standards with goods can be qualitatively ranked
(Helpman and Krugrnan, 1985). However, such comparisons are subjective: indeed
another country might have the exact opposite interpretation — it prefers the good that
does not possess certain characteristics, and considers having a certain attribute as
inferior. For example, in the US. beef is butchered so that many cuts contain parts of
more than one muscle; in other parts of the world the tendency is for different cuts to
contain only one muscle per out. To avoid using normative labels to describe a standard,
we use “Truth” and “Beauty”.26 Finally, we note that this assumption is not critical to the

results: any formulation in which the home country has a greater relative preference

 

2’ In the text words “standard”, “quality”, “attribute”, “characteristics”, and some specific examples are
used as synonyms, and shall be understood in a broader connotation.

2’ Arr opposite of vertical differentiation is horizontal differentiation, in which goods are depicted as
different without better/worse comparisons. Examples of horizontal differentiation are electrical plugs
across the globe, three (incompatible) video program formats - NTSC, PAL, and SECAM, etc.

26 Although "Truth " and “Beauty" labels are borrowed from the physics literature on quarks, it is useful to
think of traditionally produced foods as visually appealing(e.g., blemish free), and the Truth label as
representing food produced with no transgenetic techniques and as few pesticides as possible.

53

(marginal rate of substitution) for one attribute than does the foreign country suffices to
drive the model.

To pin down this modeling idea, consider the case of food safety as the attribute.
Different countries may interpret food safety differently: France may feel that irradiation
leads to safer foods, the US. may not; the US. may feel that genetically modified foods
require less pesticides and hence are safer, the UK. may feel differently. Each level of
food safety requires some fixed cost. Each country prefers to consume a broad variety of
foods. Every time a new variety is introduced, the quantity produced of each variety
declines and the cost of production rises, but the consumers gain from greater variety,
ceteris paribus. Once trade opens, each country produces a set of varieties, and wants to
import others with similar levels of safety. Love for variety is present in the model;
however, love of variety for food safety is missing, i.e., consumers in each country want a
certain level of food safety across all varieties of food they consume. Thus, variety within
food is desirable, while variety within the attribute of safety is not. Trade is driven by the
love of variety (and potential economies of scale); it is limited by differing food safety
restrictions. The benefits of free trade are greater if there is a way to standardize food
safety requirements among countries.

The model presented is static, and instantaneous adjustments are made when
needed. Countries have equal access to technologies, and differ only in consumer
preferences. All agents have perfect information. Issues related to asymmetric and hidden
information are assumed away. Consumers are able to distinguish between qualities at no
cost to them if standards for like products are in place. There are neither lobbying,

government intervention, nor network effects present in the model. Transportation and

54

standards are costless. We will think of the differentiated goods as food, but the model
formulation is general enough so that these goods could include computer chips, cell
phones, automobiles, or any other good for which consumers prefer to have the good
meet certain standards or regulations. Assume a large number of potential varieties. Firms
in the model enjoy free entry and exit, and consequently do not earn any excess profits,
and government does not impose any taxes. Since due to zero profit assumption
producers are indifferent as to what standard is demanded from them, the standard is set
at a level determined endogenously by the consumers’ utility maximizing decision. The
“consumer decides” model seems to be better suited to describe growing role of
consumer concerns in trade, although current WTO policies still reflect more of the
producers interests while discounting consumer requests due to differences in
concentration of interests.

Each country, possibly of different size, consists of like individuals, has access to
the same production technology, and differs in preference for certain attributes (in a way
that will be made precise momentarily). We start with a basic Krugrnan model where

goods are differentiated in the variety space only. Each consumer has the utility function:

1
N 5
(3.1) U = 2 (df)0 where
j=l
(3.2) 6 =[l—i],a>l
a

0 represents elasticity of substitution between varieties, assumed to be constant across
pairs of varieties; d denotes consumption of the differentiated goods; and the subscript j

implies the variety. The consumer derives utility from a large number of varieties

55

(indexed one to N). With scarce labor and positive fixed costs, it may be that only it < N
varieties are actually produced. All varieties enter the utility function symmetrically. No

capital is present: income consists entirely of wage earnings. The entire stock of labor is

used in production; consumers do not derive any utility from leisure. The demand for the
differentiated good is obtained through maximization of the CES utility subject to the

budget constrained by ownership of one unit of labor and wage normalized to one:

1
N 9 5
(3.3) Max 12:: (‘11)
N N

Assuming an interior solution, the first-order conditions (FOCS) are:

1—6

<34) (20109) 0 (d:)"“=¢m
i=1

N
(3.5) 2 Pidi =1
.=1

where¢ is a Lagrange multiplier. Solving the F OCs yields the demand for variety 1':

’0'
(3.6) di " pl

_ n 1-
2% 0
.=1

0 represents the elasticity of substitution between pairs of varieties of the same

product (i.e., food). In a symmetric equilibrium, pi = pVi :2 d,- = d‘v’i , and it is optimal

56

to purchase all varieties available in equal quantities. The price elasticity of demand faced
by producers is:

1-0'
(3.7) a + —p—"——(1 —0')

" 1
ZPi-U
.21

As it (the number of actual varieties produced”) is large, we make the commonly-
used assumption that the firm disregards the second component in the elasticity term, and
considers a to be the elasticity of demand it faces (Helpman and Krugrnan 1985).

Up to this point we followed Krugrnan (1980). Now assume that each variety can
have one of the two attributes (Truth or Beauty). Consumers have preferences over the
two attributes that imply the two are perfect substitutes”. In particular, we might view

quantities of the variety in “quality-adjusted units”, so that

(3.8) dj=dJT+2Cdf

d JQ is the quantity of variety j that possesses quality Q, for Q = T, B and A C (C =

Home, Foreign) is a “discount” parameter that marks down the physical quantity if the

variety consumed has an attribute different from the preferred one. Each country puts a

different weight on different quality: Home prefers to consume Truth, so AH 6 (0,1).

Foreign prefers Beauty goods relatively more than do Home consumers, so 2F >1.

 

27 In the Helpman and Krugman (1985) terminology actual varieties produced form a set of available
varieties with finite price — varieties not available are considered to have an infinite price.

2’ The presence of the linear subutility embedded in the love of variety CES function separates this case
from the iceberg transportation cost. In the later home and foreign varieties are consumption variables with
an elasticity of substitution of 5 in the CES subutility.

57

The linear lower tier sub-utility function where differing qualities of any variety
are substitutes with infinite elasticity of substitution between them ensures only one
quality of each variety will be consumed. The knife-edge case where both qualities are
consumed is not equilibrium. Think of consumers constructing their own consumption of
variety j by purchasing the two different components and putting them together. Using
this interpretation, the price of a unit of variety j is the minimum cost of creating the

7."
J

B

and p j represent the prices of the two

variety from the two component parts. Let p

components, then price of the bundle is p j = min( p 1 , p JB / ,1 C.) This follows from the

fact that it requires 1/ A C units of the good with quality of Beauty to provide the same
utility as one unit of the good with the quality of Truth. If p 1 < p 1/2 ,then

consumers only demand the quality of Truth for variety j. If the inequality is reversed,
they demand only the quality of Beauty for variety j, and if the two are equal, consumers
are indifferent. Since the process of the constituent components can, in principle, vary by
variety, it is possible some varieties will be constructed of only Truth, others of only
Beauty.

Increasing returns are internal to firms - an initial outlay of labor (“fixed cost”) is
needed to start up production, resulting in decreasing average cost; marginal costs are
constant within a variety — quality combination. Firms are symmetric. The presence of
scale economies ensures that in equilibrium only a finite number of varieties are
produced, each firm produces a different variety, and each variety will be produced in
one standard (Truth or Beauty) only. Each firm producing up to a certain specification

faces the same cost function regardless of its location. The number of varieties produced

58

within the economy is determined by the number of firms”. Firms enter freely into the
industry. The usual profit maximization and zero-profit entry conditions apply regardless
of the standard produced, which along with consumer demands determine price and
production levels. The equilibrium in production is described by the number of firms and
the price level. Since only labor is used in the production, following Krugman (1980) we

specify the cost fimctions:

(3.9) ci(y,-,Q)=wliQ(yi)=w(FQ+y,-MQ)

where 1,9 (W) is the labor requirement to produce y, units of variety 1' with attribute Q;

FQ is the fixed labor component for producing any variety with attribute Q, and MQ is the
marginal labor component for producing any variety with attribute Q30. Again, letting the
wage equal one, we set marginal revenue equal marginal cost to find the profit-
maximizing price charged for a variety with quality Q:

(3.10) p19 =i—1MQ

As all varieties with a given quality face the same marginal cost, the price of all varieties
with the given quality is the same. By free entry, profits are driven to zero, so that price
equals average cost. The ratio of prices equals the ratio of marginal labor requirements:

T T
(3.11) L=91§
p

be
3

 

2’ We choose to ignore integer issues when discussing number of firms in the economy.
3° Since wage in the model is normalized to one, labor requirements and costs are identical, and will be
used interchangeably in the text.

59

After solving the model, this implies that each firm will hire FQa workers (details are
provided in Table B] in Appendix B). If the amount of labor available is L, and all firms

produce the same quality, the total number of varieties that embody attribute Q will be:

L
(3.12) ng =—
FQO'

If mixture of Truth and Beauty made varieties is produced (technically possible

when consumers are indifferent between Truth and Beauty), then:
(3.13) nBFB+nTFT =£
a

From Equations 3.10 and 12, it is evident marginal labor requirements influence
the price of the varieties with different standards, while fixed costs affect the number of
varieties produces in the economy.

The discussion of the within-variety production possibility frontier (PPF) is
possible only for the case of identical number of varieties across standards. Otherwise the
economy-wide PPF would have to account for the trade-off between attributes caused by
different number of varieties across standards. From Equation 3.12, if fixed costs are
identical across varieties, number of varieties if the entire stock of labor is directed into a

single quality is the same. Figure 3.1 graphs the PPF for any variety j.

60

..................

 

 

 

 

 

 

i; li-F
.4, T ....................
S, M 1,—2F
............... T
A.... M
MB
.13
MB ‘
8 Beautv '

 

Figure 3.1: PPF in the Standard-variant Marginal, Constant Fixed Labor Requirement

Case

1/ n units of labor are used to produce each variety (It = F 0' ). The PPF (curve

ABCD) is discontinuous at zero imits: if any variety is produced in one specification only,
it uses up the labor that would have been used otherwise to cover fixed costs of producing
the “rival” quality. Due to symmetry the same holds for all varieties. If the number of
varieties in the economy is the same regardless of the standard, the prevailing standard is

chosen based on within-variety MRS (2.) and the slope of the price line (Figure 3.2).

61

 

 

 

< : Truth prevails

 

 

_ _ _ _ .. Indifference curve withl}.H

> : Beauty prevails

 

 

 

 

Indifference curve withl). H

Figure 3.2: Welfare-maximizing Choice of Standard in Standard-variant MC Case

On Figures 3.1 and 2, the slope of the AD line (connecting points on the PPF
where all labor is devoted to one standard and fixed costs are not duplicated) coincides
with the slope of the price line. Because of the fixed cost and perfect substitutability of
qualities of the same variety in the consumption, only one quality specification will
prevail for each variety. Assume this would not be the case, and some varieties would be
produced in both specifications, resulting in an allocation on the continuous part of the
PPF (line BC). Due to fixed costs the PPF is discontinuous, and choosing one standard
increases the utility available from any single variety. The opportunity cost of producing
two different versions of one variety is production of not-yet-produced varieties. Utility

maximizing standard in any variety consequently prevails for all varieties. Again, assume

62

this would not be the case, and a representative consumer consumes a mixture of Truth-
and Beauty-made varieties. Then the utility gained from the consumption of some
varieties is not maximized, and will be higher if the linear subutility-maximizing standard
prevails in all varieties.

Figure 3.2 graphically shows possible cases between within-variety MRS and the
price line theoretically discussed earlier. If the absolute value of the within-variety MRS
is less than the absolute value of the slope of the price line, Truth prevails as a standard of
choice. If the MRS is less than the slope of the price line (by definition 1.” < 1), Beauty
prevails. The solution is indeterminate if the MRS equals to the slope of the price line
and, consequently, some varieties could be made in Truth while others in Beauty. Thus,
the decision on quality is a corner solution determined by a relationship between ,1 and a
function of the cost differentials. In autarky countries produce a single quality standard as
in Krugrnan (1980).

If fixed and marginal labor requirements differ across standards, a tradeoff
between number of varieties and price occurs. In this case the consumer compares
indirect utility functions (giving maximum utility as a function of prices, income, and
number of varieties) attainable with different consumption bundles (Appendix B.2)

“Opening trade” in a one-factor model coincides with Krugman (1980), and
translates into a larger resource pool, as well as larger product market to supply. Constant
elasticity of demand assures the price levels (outcomes of the profit maximizing
conditions) do not change, the entire increase in the stock of labor in the integrated
economy is directed into production of varieties not existing in autarky, and the amount

of labor directed to existing varieties does not change. The growth in the size of the

63

market measured as an increase in the labor force does not influence the individual firm’s
output, but divides it among larger number of consumers, resulting in lower per capita
consumption of any variety. Intuitively the basic Krugman model claims that trade is
good as it (at least) increases the number of available varieties, as it is the case in this
model, and consumers in the integrated economy benefit from a larger number of
varieties available to them at the price identical with the autarky price. Therefore, having
more varieties available in an open trade than in autarky is a sufficient reason to trade and
integrate (Krugman 1980). Size — and potential increases — in the labor force enter
consumer’s utility through number of varieties available. With a large number of possible
varieties, assume each country is producing different set of varieties.

Disregard problems related to conformity assessment procedures, i.e., each
country recognizes each others’ procedures, and does not require double testing and/or
certification (e. g., even when the standards are the same). Assume Truth at Home and
Beauty in Foreign are scientifically justified on the WTO grounds, and not imposed only
as a technical barrier to protect domestic producers. In case harmonization is conducted,

it is costless.

3.2. Imposed Harmonization Potentially Causing Loss of Welfare

Common standards across countries are often treated as trade and welfare
boosting panacea prohibiting application of country-variant standards as trade policy
tools advantaging domestic producers, and consequently eliminating trade disputes.

While part of the argument — that countries with similar standards often lose (one of the)

grounds for a trade dispute — holds, such quick fix solutions might ignore country-variant
characteristics such as preferences or cultural differences. The section builds on two
widespread concepts: (1) welfare improving trade; and (2) superiority of common
standards to country-variant standards. We show when a common standard is imposed on
countries (“we will not trade with you unless there is a common — meaning our —
standard”), the joint standard might reduce welfare for one nation, autarky will be

preferred to “harmonized” free trade, and forced harmonization might impede trade.

3.2.1. Autarky

We demonstrate welfare worsening harmonization on the most simplified version
of the model. Assume marginal and fixed costs are the same across qualities: this
assumption will allow us to focus on standards and harmonization without being
distracted by price and numbers of varieties considerations.3| From Equation 3.10, if
marginal costs are the same across quality — variety combination, then prices are the

same:
(3.14) J =pB =p
From 3.12, if fixed costs are identical across standards, and the entire stock of labor is
directed into one quality only, number of all Truth and Beauty-made varieties is the same:
(3.15) nT=nB=n

For a single [autarkic] economy with identical consumers, only Truth is produced

if 2 <1 (by definition this would happen in the Home country). Only Beauty is produced

 

3' Main points from a generalized case of standard-variant fixed and marginal costs are summarized in the
Appendix 8.1.

65

if A > 1 (case of the Foreign country). Consequently, the indirect utility function if only
Truth is produced is:
1:2

6

(3.16) VT = n

.1.
P
If only Beauty is produced, the indirect utility becomes:
:2
(3.17) VB = i n 6
P

the superscript serves merely notational purposes, T indicates only Truth is produced and
consumed, B indicates only Beauty is produced. For the autarky purposes we dropped the
superscript on 2.

Define W (,1) to be a social welfare as a function of A. Producers earn zero profits,
and social welfare consists of the consumer’s indirect utility function ( Equations 3.16 or

17). On Figure 3.3 we graph social welfare attainable with autarky number of varieties

(n), applicable for both Home and Foreign countries.

We) ‘l

1-6
6

 

l
—n
P

 

----------d

 

l"
5
>2.

Figure 3.3: Social Welfare as a Function of A: Autarky.

66

The shape of the social welfare on Figure 3.3 is determined by the choice of
quality standard: if ,1 (the weight put on the consumption of the Beauty good, MRS
between Truth and Beauty) is from the (0, 1) interval — that is country values the
consumption of Truth more that it values the consumption of Beauty, country chooses

Truth as a prevailing standard. In that case the social welfare corresponds to Equation

1—6
3.16, and it is graphed as a horizontal line, intersecting the ordinate at—l-n 9 . If 2. >1,
P

country chooses Beauty as a prevailing standard, and consumes all varieties in Beauty-

make. Social welfare, represented by Equation 3.17, is graphed as a line with a slope of

1—6
—1-n 9 . Thus, the graph of a social welfare as a firnction of within-variety MRS is a
P

line kinked at A = I (the indifference point).

3.2.2. Unilateral Harmonization

Open the economics for trade, and allow for unilateral harmonization. “Unilateral
harmonization” occurs when autarky welfare maximizing standards differ across
countries and one country harmonizes its standard with the other country without a
reciprocity clause if such a harmonization yields higher utility than autarky. In the
exposition we assume no side payments or other benefits (i.e., political and/or economic
extemalities) from the country which standard is considered for harmonization by the
harmonizing country are present in the model. If the harmonizing party finds unilateral
harmonization beneficial based on comparisons of autarky and integrated social welfare,
it adopts the standard. Otherwise it vetoes the harmonization.

67

Allow countries to vary by size. From Equation 3.12, labor endowments impact
the number of varieties available in the economy. Define n to be number of varieties

produced at Home, and n* to be number of varieties produced in the Foreign country. It is

possible n at 11* due to differing endowments across countries, not different technologies

across countries. Due to the assumptions imposed (constant elasticity combined with

standard-invariant fixed costs), total number of varieties available in the integrated

economy is the summation of number of varieties previously available in autarky.
Assume countries trade only if a common standard is in place. The indirect utility

fimction if all goods are standardized to Truth is:

(3.18) VT =i(n+n*)1%p_

P
Likewise, if trade is permitted and all good are standardized to Beauty, the indirect utility

function becomes:
(3.19) VB=-)'-(n+n*)l‘—5q

P
On Figure 3.4, looking at the trade from the Home country’s perspective, we add graphs
of social welfare of harmonized trade to the autarky baseline. The autarky line is identical
with autarky line from Figure 3.3. When the countries are identical in terms of their
endowments, Home autarky social welfare function coincides with the Foreign. Equation
3.18 — social welfare when all goods are standardized to Truth — corresponds to a

1—6

. . . . . 1 ——
horizontal lrne 1ntersect1ng the ordrnate at —(n + n *) 9 .
P

68

Trade standardized

W0.) A to Beauty
Autarky
6 / (Home)
1 _
1'01 + n *) e
p Trade standardized
1.. to Truth

Q

"cl-—

 

g, 1.0 1 2

Figure 3.4: Social Welfare as a Function of 1.: Autarky, Standardization to Truth and

Beauty, from the Home Country Perspective

Equation 3.19 — social welfare when all goods are standardized to Beauty —
corresponds to a line originating at zero (when all the goods are standardized to Beauty,
and the weight assigned to consumption of Beauty made varieties is zero, total social

1-9
welfare is zero) with a slope of 1(n + n '07. The autarky social welfare function
[7

intersects each “all-standardized” welfare function once, and effectively divides the
abscissa into four intervals:

1- (0,4) - autarky dominates trade standardized to Beauty;
2. (A, 1) - Standardized trade to Truth dominates standardized trade to Beauty which in

turn dominates autarky;

69

3. (1,2) - standardized trade to Beauty dominates standardized trade to Truth which

dominates autarky; and

4. (Zoo) - autarky dominates trade standardized to Truth.

Thus, if/l < 11 , trade standardized to Beauty is worse than autarky; if A > Z , trade
standardized to Truth is worse than autarky. & , the intersection of autarky and trade
standardized to Beauty social welfare functions, is solution to:

1-0 1 1:9 1'9

A. __ n —
3.20 — + "' =— 9 :>,t= 6
( ) 1201 n ) 6 p" ‘ (MM)

 

2L , the intersection of autarky and trade standardized to Truth social welfare, solves:

1—6

1-9 9-1
(3.21) i(n+n"‘) = 9
P

if“ =Z=( " )6
n+n*

 

his»

To simplify notation, we follow the usual convention, and leave all Home country

parameters “plain”, and label all Foreign country parameters with a star (*): thus,

21H = ,1 and [IF = A *. Figure 3.5 shows the unilateral harmonization case from the

perspective of the Foreign country. Based on the Foreign parameters, we derive the

unilateral harmonization cutoff points, 33,2 * (similar to Equations 3.20 and 21).

70

Trade standardized

 

 

 

 

 

WV”) A to Beauty
Autarky
(Foreign)
1-9
i(n + n (1)7
p Trade standardized
1-9 to Truth
int—9—
p 5'
L
0 _ >
Z». * 1.0 ,1 t l *

Figure 3.5: Social Welfare as a Function of A: Autarky, Standardization to Truth and

Beauty, from the Foreign Country Perspective

If both countries have values “close” to the indifference point of one, i.e. if

{57: A 4 *} < {A A A. *} < {E A :1- *) , then standardized trade is welfare improving compared

 

to autarky for each country. The actual location of cutoff points in both countries depends
on the elasticity of substitution between varieties”, and the compensation provided by
consumption of larger than autarky number of varieties.
While we consider the case of small vs. large country in detail momentarily in
Section 3.2.2.1., we note that increases in the number of varieties available between
autarky and free harmonized trade determine the interval over which a country is willing
to harmonize: if the increase in the number of varieties from the Home country

perspective is large (n* >n), the Home country will find harmonization beneficial on a

 

’2 Elasticity of substitution between varieties is the same between the varieties of the same quality, as well
as “opposing” qualities.

71

larger interval by shifting the cutoff points (4,2) out. At the same time, the Foreign
country finds harmonization to any standard beneficial on a small interval, as 4*,2 * get

closer together.
Bypassing the assumption made earlier on Home preferring Truth and Foreign

 

preferring Beauty, we now analyze possible outcomes of the harmonization policy.33 In
each country the respective cutoff points divide the continuum of preference parameters
into four intervals. For an improved graphical exposition, Figure 3.6 illustrates the
welfare rankings Home and Foreign countries face on each interval (autarky, trade
harmonized to Beauty, trade harmonized to Truth), and plot them into a four by four grid
The Home country’s intervals are plotted on the ordinate, while the Foreign country’s
intervals are plotted on the abscissa. In the grid A stands for autarky, T for trade between
countries harmonized to Truth, and B for trade standardized to Beauty. As earlier,
“starred” parameters denote the Foreign country. For example, the sequence T > A > B
means that in the Home country trade harmonized to Truth is preferred in terms to
welfare to autarky, which in turn is preferred to trade harmonized to Beauty. Owing to the
production assumptions of equal fixed and marginal costs across standards, any 21 < 1

implies Truth prevails in autarky, and A > 1 implies Beauty prevails in autarky, as

graphed on Figure 3.6.

 

33 On the following graphs we show the case when both countries choose the same autarky standards is
trivial — it results in usual Krugrnan-like outcome.

72

‘21

 

 

 

M

 

Beauty in autarky

 

Truth in autarky

 

 

 

 

 

 

 

Truth in autarky

Figure 3.6: Welfare Ranking of Alternatives.

and the usual Krugrnan open trade result prevails.

B>A>T B>A>T : B>A>T B>A>T
T*>A *>B* T*>B*>A * E B*> T*>A * B*>A *> T*
B>T>A B>T>A 5 B>T>A B>T>A
T*>A *>3* T*>B*>A* E 3*> T*>A * B*>A *> 7*
T>B>A T>B>A 5 T>B>A T>B>A
T*>A *>3* T*>B*>A* E B*>T*>A* B*>A *>T*
T>A>B T>A>B 5 T>A>B T>A>B
T*>A *>B* T*>B*>A * E B*>B*>A * B*>A *> T"
V J

Beauty in autarky

Based on the welfare rankings of the alternatives in the Home and Foreign
countries in Figure 3.6, we simplify the potential outcomes (Figure 3.7). For example,
when T > A > B and T "‘ > A * > B", both countries rank trade harmonized to Truth higher
than autarky and trade harmonized to Beauty, and Truth will prevail. Agreement on the
standard is straightforward in the lower left hand and upper right hand corner of the grid:

when both Home and Foreign country prefer the same standard (either Truth or Beauty),

 

  

 

 

 

 

 

 

 

"I
N] .

 

A.
T T I X
:1: 1 " '
é ’1 A Foreign
Legend:

X - No agreement

B, I: Harmonized trade, one of the countries gets second choice in the agreement
B — Harmonized trade, both countries rank Beauty first

T — Harmonized trade, both countries tank Truth first

.7 — Uncertain outcome without further structure of the model

Figure 3.7: Agreements and Prevailing Standards, Case of Identical Endowments.

The case is more interesting when preferences across countries differ, and each
country prefers a different standard (upper left hand and lower right hand comer of the
grid, in ovals). X marks the case where no agreement is possible: each country prefers a
different harmonized standard, while harmonization to a standard different from the
preferred standard leads to a lower welfare then autarky. In this case the differences
between countries’ standards are too large to be bridged by trade, even at the opportunity

to consume larger number of varieties.

74

A question mark indicates a case when the outcome is uncertain unless we specify
further model structure. In either case presented both countries rank autarky on the lowest
pedestal. However, countries differ in their rankings of free trade: while they would
certainly benefit from harmonized free trade more then they would benefit from autarky,

they rank harmonization to their autarky preferred standard most. The model as presented

does not have the capacity to specify which country would execute unilateral

harmonization.

Lastly, there are cases when trade is beneficial -— each country ranks some
harmonized trade over autarky; however, one of the countries gets second choice in the
agreement: for example, Home country ranks trade harmonized to Truth over autarky, but
finds trade harmonized to Beauty welfare worsening. At the same time in the Foreign
country autarky is inferior relative to both trade harmonized to Truth or Beauty, with
Beauty topping its welfare choices. In this case world trade harmonized to Truth prevails:
the Home country vetoes trade harmonized to Beauty, as it would leave it worse off than
autarky. The Foreign country is still better off than it would be in autarky, but worse off
than it would be if the trade was harmonized to Beauty. Referring to Figure 3.6, instances

when one of the countries “gets its second best” are double underlined in Figure 3.7.

3.2.2.1. Large vs. Small Country Harmonization

Now consider the case of two countries of different sizes, and assume it (number

of Truth varieties in the Home country) is small compared to 11* (number of Beauty

varieties in the Foreign country). We depict their respective autarky and harmonized open

trade social welfare relying on Equations 3.18 and 19 on Figure 3.8. Assume in the small

75

country MRS between different qualities of the same variety is less than one, and

therefore in autarky it prefers Truth. Large country prefers Beauty.

 

A
W61) Trade standardized
to Beauty Autarky,
la‘rge country Trade standardized
1-9 " ( “mg“ ) to Truth
— (n + n ’) e

1:2
n6

 

Autarky, small country
(66H0me99)

   

 

i
p

 

 

0

V

is 4L 1.0 IL 1.

Figure 3.8: Social Welfare as a Function of A: Autarky, Standardization to Truth and

Beauty. Differently Sized Countries, Home Modeled as Small (not drawn to scale)

Notice that the length of the interval over which the small country finds the

unilateral harmonization beneficial is large, while the interval over which it is favorable

for the large country

to harmonize is narrow. In addition, if the large — Beauty preferring

—country decides to harmonize to Truth, the gain in social welfare from consuming open

trade number of varieties is small compared to welfare gains the small country would

76

 

‘zl

experience if it decided to unilaterally harmonize to Beauty. Therefore, the small country
adopts Beauty.

Figure 3.9 transposes attainable welfares in Home and Foreign countries into a
matrix of welfare rankings over harmonization intervals. Compared to Figure 3.7, when
both countries exercised the same negotiation power (measured by the size of economy,
and consequently in terms of varieties a country brings to the market), the lengths of the

intervals over which countries find harmonized trade beneficial to autarky change.

 

 

 

 

 

 

A 1
Homer X E; B B
11 i
I ? B 3
, .................................. .....................................
T T 7 B
4
T T; T X
if: 11 2* hForeiZn
Legend:

X — No agreement

,8, I_— Harmonized trade, one of the countries gets second choice in the agreement
B — Harmonized trade, both countries rank Beauty first

T — Harmonized trade, both countries tank Truth first

.7 - Uncertain outcome without further structure of the model

Figure 3.9: Prevailing Standard: Home Modeled as Small, Foreign as Large Country.
77

 

17-J-

 

3.2.3. Non-harmonized Trade

Suppose now non-hannonized trade is allowed, for example facilitated via a
mutual recognition of standards and procedures agreement. The Foreign country
produces (and trades) only Beauty (regardless what Home does). Each consumer
consumes number of varieties available in autarky, as well as varieties produced in the

other country. The Home indirect utility function from consuming both Truth and Beauty

made varieties is:

VTB =-1— n+n*ll-9

(3.22)
P

where n is a number of varieties produced at Home, and n* is the number of varieties
produced in the Foreign country. The shape of the social welfare function in Equation

3.22 depends on the value of 6. If 6 > 0.5 , the “non-standardized” trade social welfare is

convex for any ,1 6 (0,1) (Figure 3.10). If 6 < 0.5 , the “non-standardized” trade social

welfare is concave for}. 6 (0,1) (Figure 3.11).

In either case (Figures 3.10 and 11) non-standardized trade is welfare improving

to autarky. If 11 is zero, consumption of imported varieties does not add anything to the

consumption of the autarky number of varieties, but the autarky welfare - unlike in the

harmonized case (Figure 3.4) - remains protected.

78

 

‘ I

 

 

 

 

W01) (1 Non-standardized trade
Foreign produces ONLY Beauty
Home produces ONLY Truth
1 - 6 Autarky
-—(n + n *)T .
:2 s
1,, 0 j
P l
1.0 A?

Figure 3.10: Social Welfare as a Function of ,1: Autarky, Non-standardized Trade,

 

6 > 0.5 .
Non-standardized trade
WW “ Foreign produces ONLY Beauty
Home produces ONLY Truth
1 6 Autarky
j ‘ .........................................
— n + n "' g 1
p( > 3
1:32 s
.1. n e I
P 1

 

 

V

p—o
0
As

Figure 3 .1 1 : Social Welfare as a Function of A: Autarky, Non-standardized Trade,

t9<O.5.

79

 

.-
3|

If the assumptions regarding the size of marginal costs across varieties were
released, different prices would result across varieties. Differences in country size or

cross-country differences in fixed costs will change n and n*. Effectively they would
change the “critical value” (that is, i and :1- : where consumers are indifferent between
Truth or Beauty), and shift curves. Nevertheless, the general conclusions would remain

the same.

3.3. Large Number of Countries

In the two-country world we have shown unilateral harmonization can be welfare
worsening depending on the within-variety MRS, while non-standardized trade is always
beneficial. We now extend the analyses to a larger number of countries. Assume there are
(t+b) countries in the world: t countries prefer Truth standard to Beauty (2. < I), b
countries prefer Beauty attribute to Truth (A > I). We allow fort at b. Countries of each
type are assumed to be of the same size: each of the t countries produces n varieties, each
of the b countries produces n* varieties. Therefore, the number of varieties (ignoring the
quality aspect) consumed in autarky by any single “t” country is n (or n* in any “b”
country if their endowments of labor differ), the number of varieties consumed if all
countries with the same preferences for a certain attribute would form an agreement is tn
(or bn *), and finally the total number of varieties produced in the world is (tn+bn *). The
generalized notation is introduced to explore potential market power.

We define a club as a set of countries with similar preferences clustered in a trade

agreement. Formation of agreements is costless. The autarkic standard prevails in each

80

bloc, and essentially becomes a club good. By recognizing the same standards within a
club disputes are likely to be prevented. Non-members (or members of a different
agreement) do not use the excludable standard of the club, and do not trade with the club
members. Therefore, groupings of countries become de facto clubs in a Buchanan sense
(Buchanan 1965), where standards have the characteristics of club goods, and non-
members are excluded from consuming the benefits — that is unrestrained trade within
club.

Assume countries in the model form a Truth club and a Beauty club based on their
preferences. In the Truth club — formed by countries with A < 1 , and thus preferring Truth
to Beauty — each member has access to and consumes tn varieties. The indirect utility
function of a representative consumer of a Truth club (TC) is:

(3.23) VTC = 1(2):)?
P

Similarly, each member of a club formed by Beauty prefening countries (A > 1)
has access to bn" number of varieties. The indirect utility function of a representative
consumer in a Beauty club (BC) is:

(3.24) VBC = law)?
P

As earlier, we graph the attainable social welfare within a club as a function of
marginal rate of substitution (2), and compare it with autarky social welfare (Figure 3.12).
Social welfare attainable within a club exceeds the autarky social welfare over the entire
interval: clubs allow consumption of larger-than-autarky number of varieties at autarky

preferred standard.

81

W01) .1 Club

    

 

Autarky

 

 

p------ -----------d

 

5"
Q
3.

Figure 3.12: Social Welfare as a Function of ,1: Autarky, Club.

3.3.1. Non-standardized Trade between Clubs

If non-standardized trade between clubs was allowed, each country would have
access to (tn+bn *) number of varieties. The total social welfare attainable would be:

1—6

9 e

(3.25) V=-1- m+bn*21-9
10

We compare non-standardized free trade with club and autarky welfares. Graphically the
scenario is illustrated on Figure 3.13 (for6 > 0.5) and 3.14 (for6 < 0.5 ). As in the two-
country case (Figures 3.10 and 11), non-standardized trade is welfare improving
compared to autarky, as well as club trading. However, in the presence of clubs formed
by multiple countries with the same preferences, the lowest attainable social welfare
when consumers completely discount the consumption of the other good coincides with

“club autarky”, rather that autarky per se.

82

World Non—standardized trade

W0.) A ’1'
1’ Clubtrade

 

 
 

 

‘ul—

 

.------ ----------------k.------

 

Ml

p—a
C

Figure 3.13: Social Welfare as a Function of A: (t+b) Countries, Autarky, Non-

standardized Trade, 6 > 0.5 (not drawn to scale).

83

World Non-standardized trade
W61) A r ,'
’ ’ Club trade

 

 

-]p-(tn + bn *) 9

 

.------ --------..-------\.--------.

 

 

y—n
O
s."

Figure 3.14: Social Welfare as a Function of 21: (t+b) Countries, Autarky, Non-

standardized Trade, 6 < 0.5 (not drawn to scale).

3.3.2. World Standardized Trade between Clubs

Now assume non-standardized trade is not allowed, and instead countries seek to
pursue the harmonization route. Club member are homogenous in all aspects (assumption
of identical MRS between attributes might not suffice), and none of the members exhibits
any selfish behavior (i.e., they negotiate as a club, see the EU example). The indirect
utility function if all goods are standardized to Truth is:

1-6
(3.26) VT = inn + 6m)?
P
Likewise, if all goods are standardized to Beauty, the indirect utility firnction becomes:

84

1—6
(3.27) VB = inn + 6m)?
P
Figure 3.15 graphs social welfare attainable with (tn+bn *) number of varieties — but

harmonized to one standard - together with club and autarky welfares.

 

A
W ,1 W ld trad .
U haglrronizecf to 1’ Clllb trade
Beauty I’ I
1 1— e ’
11

70" + b" ) 9 1’3 World trade

[I harmonized to

l Truth

1 1:2
—p—(tn)g 1.....-—

 

3
3|

 

"cl-—

 

 

 

1

>3 ll __._.

1.0

A

Figure 3.15: Social Welfare as a Function of 1: (t+b) Countries, Autarky, Standardized

Trade (not drawn to scale).

Figure 3.15 represents the comparison of autarky, club and harmonized world
trade social welfares. In the absence of club trading (i.e., if it was prohibited or otherwise
prevented), any individual country would be willing to harmonize (with the exception of
very low values of marginal rate of substitution) to either Truth or Beauty. Presence of

club trading narrows the length of the interval over which harmonization to a common

85

world standard is beneficial (i3) : while on this interval either of the harmonized social

welfare functions dominates club welfare, in case an agreement on common standard is
not reached, countries are still better of in clubs compared to autarky.

As in the earlier case, size of the clubs — measured by the number of varieties they
cumulatively provide — in comparison to the total number in the world determines the
club’s willingness to harmonize its standards with the other club. If the club is small
compared to the rest of the world — i.e., it is formed by a smaller number of countries,
countries are smaller in terms of endowments — by adhering to a certain club standard
consumers would be restricting their welfare. On Figure 3.15 size of the club is
represented by the position of the “club” social welfare compared to harmonized trade
social welfare functions: if the club is relatively small compared to the rest of the world,
the world trade social welfare function when all trade is harmonized to Truth shift up,
while the world trade social welfare function when all trade is harmonized to Beauty

pivots to the left around zero, as outlined by the arrows. Resulting is a large interval

(£7) over which a club is willing to accept other club’s standard to experience larger

welfare. Again, for extreme values of MRS — i.e., is a country puts small weight on the
consumption of the other standard, a club autarky yields higher welfare than harmonized

world trade.

86

3.4. Core Analysis

We now turn to investigate whether or not the outcomes of the models presented
are in the core. For the purposes of this chapter, we define the core as a set of allocations
that no subset of countries would find separation from the other countries and trade
among the subset beneficial. The same applies for the two country case: the core is a set
of allocations when no country would find deviation from the agreement and consequent
autarky beneficial.

Non-harmonized trade, if allowed, is in the core on the entire continuum of 1: no
set of countries can be made better off by segregating and trading among themselves
regardless what relative weight they put on the consumption of Truth and Beauty made
varieties.

The outcome of the harmonized trade case depends on the position in the matrix
of choices in Figure 3.7 (or fitted to the multiple country case). By definition, the
allocation when both countries prefer harmonized trade to the same standard (both
countries prefer Truth in the lower left hand comer, and both countries prefer Beauty in
the upper right hand comer) — there is no allocation that yields a higher welfare, and no
country can be made better off by a different allocation.

When countries in autarky choose different standards we focus on the upper left
hand comer where Home prefers Truth, and Foreign Beauty. The lower right hand
corners mirrors the analysis. When the Home country values consumption of Beauty very
little and the Foreign values consumption of Truth very little, no agreement is possible
(marked X on Figure 3.7). The multiple country equivalent translates into formation of

two sub-agreements, each trading at its preferred standard. As there exists no other

87

allocation that would allow a subset of countries further to segregate and trade, the
trading arrangement consisting of two clubs is in the core. A global club arrangement, if
formed, would not be in the core: as given the countries’ preferences, harmonized trade is
welfare worsening, and each subset of countries will be made better off by trading among
themselves.

When the differences between countries’ valuation of standards exists, but those
differences do not prevent them from forming an agreement, such an allocation is in the
core, even though one country (or one set of countries in the multi-country case) gets its
second choice of standard in the agreement (marked by double underlining in Figure 3.7).
Separation from the (global) agreement and within club trading would not yield higher
welfare. The same applies for the cell labeled with a question mark when both (sets of)
countries prefer some harmonized trade to autarky: assuming they would reach an
agreement on the prevailing standard, a deviation from the agreement does not result in

an improved distribution.

3.5. Applying Lessons from the Club Theory

Using the vocabulary of public economics, the country providing the public good
of the standard (or unilaterally harmonizing) finances its provision via discounted utility
it obtains from the consumption of more varieties. The country to whose standard
unilateral harmonization took place de facto “free rides” on the other country - it benefits
from a larger number of varieties it originally preferred without changing its behavior. If

the weight put on the consumption of a less preferred standard is too small to justify

88

unilateral harmonization or both countries would benefit from unilateral harmonization
but are adopting a waiting strategy (cell marked by question mark in Figure 3.7, fitted to
the multi country case), a global agreement is not feasible. The only way to take
advantage of (at least some) gains from trade in form of larger number available varieties
is for the countries to resort to sub-global agreements among countries with similar
preferences. The number of varieties consumed in a sub-global agreement exceeds the
number of autarky-consumed varieties, but is less than the number of varieties consumed
if a global agreement was formed. Countries stay in a “witlrin-club” autarky: trading
varieties only with countries with the same preferences, showing optimal harmonization
(or any other policy coordination) domain can be subcglobal. The current stage of the
WTO negotiation is a likely fit: despite the advantages of a global domain achievable via
unilateral concessions (“financing” the provision of a public good of standards, and
consequently, free movement of goods), countries adopt a waiting strategy and form sub-
global agreements. The emergence of sub-global agreements clustered based on
countries’ preferences towards certain attributes is supported by consistent with social
choice (Olson, 1968) and club theories (Buchanan, 1965).

If standards are common across all countries, or any country is willing to “finance
a provision of public good” via unilateral harmonization, a global club is formed. Thus,
as described above, countries sharing similar members’ characteristics (i.e., choosing the
same standards) voluntarily cluster themselves into clubs: “Truths” will form a club, as
will the “Beauties”, resembling a separating equilibrium. Otherwise, in sub-global clubs,
countries consume more varieties at their original quality standard than in autarky,

without any standards adjustment. Each firm within any club produces one variety for all

89

consumers in the club, everyone specializes, and everyone trades, but only with the
members of its own club, including global. WTO-compatible clubs34 in this paper are not
necessarily strictly the clubs in Buchanan’s sense that provide or facilitate the provision
of public goods per se. Nevertheless, in a broader sense, clubs facilitate movement of
goods within a group of countries by having only one standard prevailing within a club.
The club standard does not necessarily have to be institutionalized to facilitate trade, or to
provide a club good of free movement of goods (and services) among a certain subset of
countries. A dominant standard within a club thus serves as a sort of exclusion
mechanism keeping countries with different preferences external to the club. Non-
members are — and also want to be - excluded from the trade with members.
Theoretically speaking, the global domain — encompassing all countries - might
not be workable to reach: “unless the number of individuals in a group is quite small, or
unless there is a coercion or some other special device to make individuals act in their
common interest, [the] rational, self-interested individual will not act to achieve their
common or group interest” (Olson 1965, pg. 2). Countries homogeneous to a certain
degree form a smaller group of fewer players able to bargain more effectively and
efficiently and provide a club good (for example, grades, standards and labeling
harmonization) where provision of public goods has failed (or is likely to fail).
Homogeneity is defined in terms of similar consumers’ attitudes, industry structure,
interest groups, income, tolerance levels, perceptions of risk etc. Requiring outsiders to
produce their product up to a certain standard (or label or impose any other NTB) can act

as a tax or technical barrier to trade, and probably leads to prohibition of goods from

 

3’ Preferential agreements presented in this paper do not violate the WTO rules since they do not make
trade with non-members more difficult than before an agreement was in place.

90

outside the club. Similarly, leaders are more likely to emerge in a smaller, more
homogeneous setting (i.e. Kindleberger 1981 or Fratianni and Pattison 2001). A real
world example of agricultural biotechnology clustering countries into clubs based on
similar positions is discussed in Chapter 5.

While we only discussed a simple case of discrete standards resulting in two clubs
being formed, with a more complicated setup multiple clubs would occur, thus lowering
chances of survival of a global organization already in place -— if any. Therefore, the
number of clubs is predetermined by the existing number of standards, assuming
standards are not clustered and the marginal entrant into any club can be clearly
determined. Nevertheless, as supported by Olson, members are more likely to reach a
conclusion in a smaller grouping of more homogenous members compared to a

heterogeneous global organization.

3.6. Chapter Conclusions

The chapter built a Krugman-like model of endogenous standards setting and
consequent trade. Consumers in each country loved variety, but had different marginal
rates of substitution between different standards of the same variety. Due to the constant
elasticity of substitution, gains from trade are limited to a greater number of varieties
available. We studied a discrete version of a quality standard: that is, a variety either
contained a certain trait (“Truth”) or it did not (“Beauty”). The model abstracted from
price effects due to the assumption of constant elasticity. We considered two different

equilibrium candidates: open trade without harmonization (and free of any technical

91

barriers to trade), and open trade with harmonization. In the first case by assumption
Home and Foreign produced different sets of varieties and traded. We showed that trade
without any harmonization (equivalent to consuming mix and match of varieties, but
never any variety in both specifications) was always welfare increasing as compared to
autarky. However, when one of the countries required every variety on its domestic
market to be produced up to a certain quality standard, trade at an imposed uniform
standard could be welfare worsening as opposed to autarky depending on actual ratio of
marginal labor requirements to produce each quality, the increase in the number of
varieties available, the elasticity of substitution and the parameter “discounting” the
utility gained from a consumption of varieties produced at a different standard.
Nevertheless, we have shown autarky can dominate free trade if there is a distortion
involved. Transposing the result into the policy arena, we conclude that in some cases the
negative “harmonization effect” —- or the effect of a standard different from the one
originally preferred - exceeded the increased variety effect, and a country would veto the
harmonization and trade.

Once trade opened countries with the same autarky-prevailing standards “formed
agreements”: if the same standard prevailed across the world, a global agreement is
formed. A global agreement could be also formed when any party decides to harmonize
unilaterally, bringing in issues of public goods provision, such as financing and free
riding. Otherwise countries formed smaller agreements with their allies. Nevertheless, in
either type of agreement, each country benefited from having access to a larger number
of varieties than in strict autarky, even though the number of varieties available in the

sub—global club is smaller than it would be in a global club. The model also supports

92

some stylized facts of international trade: “regional” trade agreements being formed
among geographically distant partners based on their preferences, and “large countries
free-riding on smaller ones” due to their willingness to execute unilateral harmonization
owing to the relative larger gains that small countries obtain from trade. While the actual
process of negotiating a free trade agreement might not necessarily follow the process
outlined in this paper, it clarifies why countries choose negotiation partners with similar
preferences or willingness to pursue unilateral harmonization to secure success. Global
negotiation when countries have different preferences could be problematic. In addition,
agreements discussed in this paper do not necessarily have to be explicitly
institutionalized; rather, they can take form of an informal club.

Another stylized fact is that the small trade agreements act as a stumbling bloc in
the world trade negotiations. The model confirms that it is indeed true that non-
harmonized trade can be desired. However, when harmonized trade is desired, subglobal
agreements can be welfare improving when global trade is welfare worsening (that is, on
certain intervals).

The discussion in the paper was limited only to two types of countries with two
discount parameters (is). However, in reality the world consists of countries with more
heterogeneous preferences, and consequently the number of sub-global agreements is

likely to exceed two. This extension is a topic for future research.

93

CHAPTER 4

ECONOMIES OF SCALE: AUTARKY, HARMONIZATION AND COMPROMISE

Chapter 3 explored standards selection decisions in autarky and consequent
agreement formation based on the “love of variety” in the usual Krugrnan (1980) setting
of non-sunk fixed costs, constant marginal costs, and constant elasticity of substitution
between varieties; Chapter 3 enhanced this analysis by including a discrete attribute
(standard) to the model. Owing to production assumptions, a within-variety production
possibilities frontier was “discontinuous” with a linear continuous section (Figure 3.1).
Constant elasticity of demand kept the prices unaffected by the increases in the labor
force, and allowed exploring changes in the number of varieties available, but prevented
employing a traditional economies-of—scale argument originating in decreasing costs over
larger output. Finally, strict discreteness of standards (the product either did or did not
possess a certain attribute) did not permit analyses of a compromise or allowed countries
to make small changes in reaching a common tolerance level.

Chapter 4 follows up on the problem introduced earlier: it addresses the
differences in national standards .for like products and their potential harmonization,
either unilaterally or by defining a common tolerance level. Relying on graphical
illustrations, it releases assumptions regarding fixed and marginal labor requirements,
elasticity, and the discreteness of a standard. Standards are associated with tolerance

levels towards certain attributes, such as occurrence of genetically modified (GM)
organisms and foodstuffs (GMOs) or pesticide residues in the food system: Beauty

(complete acceptance, or firll tolerance) and Truth (complete rejection, or zero tolerance)

94

being extreme cases, with the possibility of some intermediate level (i.e., products
containing more/less than a certain amount of the attribute are labeled — either mandatory
or voluntary). The assumption of the horizontal separation between standards on a global
level, and vertical differentiation from the standpoint of consumers in individual
countries remains honored. Once again we emphasize the connotation of individual
standards: while Home considers T ruth-made goods superior to Beauty-made, Foreign
has the exact opposite inclination. Nevertheless, from the WTO perspective, Truth and
Beauty are equal in terms of performance, and as such are classified as like products.

This chapter emphasizes modifications of consumption and production under a
“traditional” economies-of-scale argument with decreasing marginal costs. While
different standards are not directly comparable in terms of their performance, their
respective labor requirements are directly comparable. Assume marginal labor
requirements are distributed along a continuum; for any level of output produced the
marginal requirement of producing Truth (“lower tolerance level” or a stricter standard)
exceeds the one of producing the same amount of Beauty (“higher tolerance level,” more
lenient standard). Sunk (fixed) labor requirements ensure the PPF is continuous (details
discussed in Section 4.2). The number of varieties available depends on the country size
(defined in terms of workers).

We start with a model description and graphical solutions to the general example
of autarky standard selection (not related to any specific sector), followed by
harmonization, and compromise between two countries with different preferences. We

analyze the two country case (one each of the two different types), later releasing the

95

equal size assumption. As in the previous chapter, total welfare considerations are limited

to maximizing a representative consumer’s utility, as firms’ profits are confined to zero.

4.1. Consumption

Assume the world consists of two representative countries: Home and Foreign,
with identical endowments of labor (this assumption will be relaxed later) and access to
the same production technologies, but with different within-variety marginal rates of
substitution depending on an attribute (e. g. percentage of the GM ingredients). Standards
are defined on the bases of attributes present as tolerance levels; and in border cases
Truth stands for zero tolerance, and Beauty for full acceptance.

The model is based on the love-of-variety in differentiated goods. There are many
differentiated goods available. The differentiated goods may be produced with different
levels of a specified attribute; one or both country(ies) may choose to subject the level of
this attribute to a standard(s). Assume there are L identical consumers in each country,
each consumer is endowed with one unit of labor earning a wage of w. Each consumer’s
preferences are defined as35 :

N
(4.1) UC = 21(1))
j =1
where d,- represent consumption of any individual variety, superscript C takes the values
H (for Home) or F (for Foreign). Due to the symmetry imposed by assigning same

qualities of different varieties of the differentiated good equal weights, the share available

 

3’ Unlike earlier, here we refi'ain from specific functional forms for the upper tier utility function, but define
the lower (within-variety substitutes) tier.

96

for the “quality variant” goods is spent evenly on each variety. In addition, v' > 0 and

v" < 0 , ensuring v is increasing and concave. Assume that the number of varieties in the
differentiated sector (N) is sufficiently large so that the budget share of each of them is
small, and the impact of a change in one price on the marginal utility of income can be
ignored. For any variety j the elasticity of substitution — and elasticity of demand facing

an individual producer is defined as:

dc ° '
(4.2) ej = ——J——"i- = {—3—} > o
dpj dj djv
Also assume
as]-
(4.3) — < 0
adj

Presenting a generalized case of love of variety utility function without placing
any limits on the elasticity of substitution (other than decreasing in consumption,
Equation 4.3) does not exclude trade for reasons of accessing a wider variety of goods,
while at the same time it allows for the traditional economies of scale argument - as will
be clarified momentarily. As in Chapter 3, assume consumers in each country are

composing their consumption from two available qualities (Truth and Beauty) such that:
. _ I." C 1.3

(4.4) dJ—dj +,1 d]

where T is for T ruth and B for Beauty, C in the superscript stands for country (Home or

Foreign). To account for differing preferences with respect to qualities across countries,

assume that:

(4.5) 0 < /1 H < 1 (preference parameter in the Home country)

97

(4.6) A. F > 1 (preference parameter in the Foreign country)
Equations 3.4 - 6 define different quality versions of any differentiated good as
substitutes in consumption, as represented by a linear subutility function, with different
weights put on the consumption of different standard across countries. Since the total
utility is increasing in each subutility obtained from a differentiated good, for the total
utility to be maximized, each linear subutility has to be maximized. Thus, the consumer is
maximizing based on the within-variety marginal rate of substitution (1C) in each country.
Mixing and matching of Truth and Beauty varieties is prevented: if, for example,
consumption of Truth-made version is chosen as welfare maximizing as opposed to
Beauty-made version, then Truth standard prevails for all available varieties. In this case
mixing in some Beauty-made varieties would decrease the utility. The number of
varieties available under Truth or Beauty is assumed to be identical regardless of the
prevailing standard, allowing us to concentrate solely on the vvithin-variety quality
tradeoff and ignore choices between qualities and number of varieties consumers would
have to choose from otherwise. Due to the symmetry imposed in the model,
considerations given only to a representative differentiated variety are appropriate.
Assume standards — if present - are introduced and imposed by respective country
governments at no cost based on consumer preferences to facilitate consumer’s decision.
Countries are not necessarily obligated to recognize each other standards, as it will be

shown in the last section of the chapter on the case of transgenetic crops and GMOs.

98

4.2. Production

L available units of labor (one unit per consumer) are divided between it produced
varieties of differentiated good. In the production of the differentiated good we assume a
hybrid production structure providing differentiated varieties in the presence of
increasing returns to scale and sunk cost“. Each variety of the differentiated sector
requires an unrecoverable sunk cost investment, defined by Tirole (1988) as “those
investment costs that produce a stream of benefits over a long horizon but can never be
recouped.” Consider the sunk costs being the cost of research and development, learning
by doing incurred regardless whether the good finally enters the market, licensing fee, or
any one-time entry fee. Sunk costs of Truth exceed sunk costs of Beauty. Consumer’s
love for variety in the differentiated product prevents the usual increasing returns to scale
monopoly outcome, and allows for a monopolistic competition model. The sunk cost
ensures that a certain (finite) number of varieties is produced, and markets are not
contestable. Once firms invest in sunk costs, they have an incentive to stay in business,
and do not surrender themselves to other firms or give the other firms an opportunity to
grow large. The differentiated sector exhibits increasing returns to scale with decreasing

marginal costs. In general,

(47) Q = f (L)

(4.8) 9'” Q = f(.9L),w >1,.9 > 0

Labor is more productive in the Beauty-made versions: for any Truth made version:

 

3’ Usual outcome on the in the presence of increasing returns to scale and sunk cost is a monopoly. Detailed
production structure would not improve the intuition gained from the model. Sunk costs, compared to usual
fixed costs, are to maintain continuous PPF and influence number of varieties.

99

T
(4.9) y [T =11.“

For any Beauty made version:

B
(4.10) yiB :1? ,where

MJD aB>aT>l

In the previous L represents total supply of labor available in the economy, while 1
corresponds to the stock of labor used to produce any individual variety. The same
production structure and standards choice apply to every differentiated variety produced
in the economy, and are the same for both countries. aT and a3 are linked to the standards
themselves -— higher a represents higher productivity labor can achieve in that particular
standard, and are assumed to be distributed along a continuum with Truth (a7) and Beauty
(a8) representing border cases.
The number of varieties produced (n) is an increasing function of country’s labor
force only:
(4.12) n = n (L)
In addition, assume that number of varieties in each country is determined exogenously in
some previous period by the unrecoverable sunk cost interpreted as research and
development expenditure. Thus, the number of varieties produced is independent from
the prevailing standard”.
When mapped into a within-variety production possibilities frontier (PPF) space,

the PPF is non-symmetric (due to different productivity of labor in Truth and Beauty

‘

37 Simplifying assumptions regarding, for example, the research and development costs, number of
Varieties, etc., allow us to concentrate on the impact of harmonization and compromise in standards across
countries, as well as set the stage for the last chapter. Indisputably, a detailed exposition of the model
Would add rigor, but not necessarily the intuition we seek.

100

versions of the same variety) and convex to the origin (due to decreasing marginal costs
resulting from increasing returns to scale in both varieties). Economies of scale observed
in the model are internal to firms, not correlated with the size of the industry, and are

apparent in any quality-variety specification.

4.3. Competitive Equilibrium vs. Social Optimum

This section discusses reasons for employing a benevolent social planner as a
decision maker rather than relying on competitive markets. The Pareto efficiency of a
Walrasian equilibrium in competitive markets depends on preferences and technologies
being monotone and convex. However, the production function described in the previous
section was non-convex with increasing returns to scale, and a sunk fixed cost was
observed by every entrant. In this case the competitive equilibrium is inefficient resulting
in a market failure due to non-convexities present while the social optimum accounting
for non-convexities and using values for social benefits and costs is Pareto optimal. In an
economy with increasing returns, the first best efficiency may be impossible to attain
through an equilibrium concept based on market prices (Dall’Olio and Vohra, 1999).
Nevertheless, the presence of increasing returns to scale does not rule out production
efficiency in the sense of producing on the PPF. Symmetry simplifies the problem to a
benevolent social planner maximizing social welfare — or utility obtained from any
differentiated good based on the relationship between the within-variety marginal rate of
substitution between Truth and Beauty made versions and the production possibilities

frontier, and avoids problems associated with average cost pricing in a decreasing

101

marginal cost environment. If the utility gained from any differentiated good is
maximized, then the total welfare is maximized as well (assuming labor force is equal
size), since producers’ profits are assumed to be constrained to zero. Nevertheless, the
main goal of the paper is in demonstrating the benefits of standard’s adjustments using
the traditional economies of scale case, rather than arguing over the Walrasian

equilibrium and social optimum in the presence of the Second Welfare theorem.

4.4. Autarky Decision

In the socially optimal equilibrium, social planners in each country allocate goods
and factors such that the consumers’ utility is maximized; firms earn zero profits; labor
and goods markets clear.38 Recall that each economy consists of L consumers, each
owning one unit of labor. Analyses are carried out on the aggregate level assuming
egalitarian distribution of welfare. Each variety requires sunken set-up costs, and can be
produced in either standard. Increasing returns to scale and substitutability in
consumption (Equation 3.4) ensure in the optimum only one standard of each
differentiable good will be produced.

The relationship between the two countries’ standards in an autarky setting can be
depicted on an indifference curve map (Figure 4.1). The PPF for variety 1' is depicted by
the curve ATAB. This curve is convex to the origin due to the assumption of increasing
returns to scale. The points AT and A 3 represent the maximum possible production of
Truth and Beauty version, respectively. Since production of Beauty is cheaper (or less

factor intensive), A 3 > A1. Each country has access to the same production technologies,

 

3‘ Planner could equally well maximize profits subject to minimum utility constraint.

102

so that the PPF is identical for each. Due to symmetry imposed in the production, the

same holds for all products. From 4.4 the home country consumer must consume

x/A. H units of the Beauty version “like” product i to generate the same utility as
generated by x units of Truth, so that his indifference curve is a line with slope —,1H. His
utility is maximized at the comer solution AT; i.e., where all production of variety i is in
Truth standard. Similarly, the foreign consumer must consume 11Fx units of the Truth
version to generate the same utility as generated by x writs of Beauty, with ,1F >1 . This
means that the foreign country consumer’s indifference curves are also linear, with slope
41F. Her utility is also maximized at a comer solution, but at the corner A 3, indicating

that at the optimum the foreign country will produce only Beauty version of variety i.

A

  

a

E

E“ H

“5 Home IC (,1 <1 )

.é’

t:

D

- Foreign IC (2‘3 1)

AT

 

- L

A 3 Units of Beauty ,

Figure 4.1: Autarky Decision in General Equilibrium

Since the two types of good i (i.e., food) are indistinguishable to the consumer at

the time of purchase, maintaining these optima requires regulatory standards. Assume

103

costless enforcement of standards. The autarky equilibrium is a corner solution, in which
Home produces and consumes at AT, while Foreign consumes and produces at A 3. That is,
the Home country chooses the Truth standard, while the Foreign country chooses the

Beauty, and all production exactly meets the respective standard.

4.5. Open Trade

Opening trade translates into a larger pool of labor available to produce
differentiated goods (and a larger number of varieties), as well as a larger product market
to supply with finished goods —- assuming the existence of different attributes is not
sufficient to prevent trade. For now assume countries are of equal size. Labor is not
mobile internationally, and disregard any problems related to conformity assessment
procedures and legitimacy of different standards across countries under the WTO rules.

With love of variety bestowed on consumers, trade between countries is driven by
both increased variety and economies of scale originating in increasing returns and larger
amount of output at a lower price. Recall that the number of varieties produced in an
economy is a function of labor available and predetermined in the earlier period by
research and development expenditures: as integrated market labor supply exceeds the
autarky labor supply, open trade number of varieties exceeds autarky number of varieties.
Also, the amount of labor available for each variety in the integrated economy increases,
resulting in a greater output and PPF shifted to the right (covered in detail in 4.5.1). We

label the increase in output y (y>1).

104

For a graphical exposition of increased number of varieties we refer to Figure 4.2.
In autarky each country produces and consumes own set of varieties. There exists a
subset of varieties in which countries have the potential to realize economies of scale,
assuming they settle on a common standard (to be elaborated momentarily). If they reach
a common ground regarding a standard, the labor savings can be used to produce
varieties (at a common standard) previously not available in the autarky. Thus both
countries realize the economies of scale effect of producing at the same standard, as well
as potential variety effect. We focus on the economies of scale effects from
harmonization to a common standard on a single representative variety from the

overlapping intersection of varieties (Figure 4.2).

Country-specific varieties with no
attainable economies of scale (local
produce, wine, cheese, cultural goods, etc.) Foreign country
varieties (act.
produced)

 

 

     

Home country
varieties (act.
produced)

   

 

Set of additional varieties
available to be produced
from the labor savings

realized by economies of
scale from harmonized Varieties with potential economies of scale effect

production when produced at the same standard and traded.

Savings from economies of scale could be used to
produce varieties not produced in autarky.

Figure 4.2: Number of Varieties

105

Open trade is thus a static decision whether or not to open the trade this period,
and at what standard. Assume costless and instantaneous adjustment to a new standard if
needed — regard it as a costless technology calibration after the sunk cost was already
made.39 By utilizing increasing returns to scale and supplying larger market, average
costs of production are reduced leading to a greater efficiency. Potential equilibrium
candidates, besides autarky are:

l. Unilateral harmonization;

2. Mutual compromise on standard (finding a common tolerance level); and

3. “No trade” - the standards are too far apart, economies of scale are not sufficient
to justify change in standards, and there are no gains from trade to be had.

Consequently countries do not engage in any trade even if it is legally and

technically possible (i.e. we do not assume autarky).

4.5.1. Unilateral Harmonization

In the starting position foreign standards — from the home’s point of view — are
more lenient: Foreign country in autarky adopted a full acceptance policy (Beauty), while
Home maintains a more labor intensive zero tolerance (Truth). Beauty consumers are
willing to consume the Truth version of the good (and vice versa) if compensated
somehow, and conversely with trade, production costs fall due to economies of scale.
There is a possibility that the reduced production costs are sufficient to induce one of the

parties to change their standard. This situation is depicted in Figure 4.3.

 

3” If change of standard was not costless, the new world PPF (graphed in Figure 111.2) would like to the left
of the graphed PPF, as part of the available resources would be used up in the adjustment.

106

L
7

—TT

Units of Truth

World PPF

 

 

 

A3 1% T B )23 TBT Units Ofr
Beauty

Figure 4.3: Decision to Harmonize

The PPF A BAT and the foreign indifference curve through A 3 start the foreign
situation in autarky. With the opening of trade all production of a certain variety takes
place in one country, the labor allocation in this variety increases; The increased labor
supply results in increased production levels MT and Mg. Additionally, a single firm
could produce all of the world’s demand for this variety and thereby capture economies

of scale. Thus the trade (“world”) PPF has endpoints Tr>yAT and T 3>yA 3. Moreover,

107

because the higher costs of meeting the Truth standard are captured in the (sunk) fixed
cost term, the economies of scale are greater in the production of the Truth-made good.
This means that the proportionate increase in the endpoint on the ordinate (Truth axis) is
greater than the proportionate increase in the endpoint in the abscissa (Beauty axis):
TH/A H > T ,/A 1.-

By definition the foreign (i.e., country starting with the full acceptance standard)
has half of the labor force, allowing it to be either at I/2TT (when producing on the Truth
standard), or A 3 (when it produces on the Beauty standard). Thus, the choice for the
foreign country is between 1/2Tr and A 3. However, 1/2Tr guarantees higher obtained
utility from variety i, and thus the foreign country decides to accept the Truth standard,
preferring more of Truth quality good in trade to less of the Beauty quality good in
autarky production. The same holds for each variety consumed, and when summed across
varieties, the utility of the foreign country increases. Due to symmetry of the domestic
and foreign indifference curves, the foreign indifference curve intersects the ordinate
below 1/2TT, but the domestic indifference curve intersects the abscissa to the left of
1/2T3. That is, the economies of scale are sufficient to compensate the foreign country to
harmonize standards at Truth, but insufficient to compensate the domestic country for
harmonization of standards at Beauty. Consequently we assume that standards are
harmonized at Truth (the case of intermediate standard is dealt with momentarily).

Thus, the Truth standard prevails and firms are realizing economies of scale from
access to a bigger market. Consumers in both countries enjoy a larger number of

varieties. Due to balanced trade, each country is trading the same number of varieties. In

108

the Truth-loving country the utility increases due to the larger number of varieties

supplied at the desired level of standard.

4.5.2. Mutual Compromise

By compromising on a standard in terms of finding a common tolerance level the
countries increase the possibility of mutually beneficial trade. Assume the T ruth-loving
country lowers its standard from T to T’ (for example, it moves from the zero tolerance
policy to ten percent tolerance level). For simplicity assume there is no change in the
Beauty-loving country. Since the cost of producing goods of quality T’ is less than the
costs of producing goods of quality T, the new autarky PPF becomes Ang. As illustrated
in Figure 4.4, with open trade the world PPF becomes TT’TB. The foreign-country
indifference curve passing through A 3 also rotates, as the change in quality from Tto T’
also changes the consumers’ marginal utility of the Truth-quality good relative to the
Beauty-quality good. Because the standard T’ is closer to the foreign standard that is T, it
takes fewer units of T’-quality goods to equal in utility one Beauty-quality good (in the
foreign utility function); consequently the indifference curve rotates counter-clockwise.

The two effects of the compromise on the Truth standard are thus to expand
outward the PPF on the T ruth-standard axis, and to pull inward the indifference curve on
that same axes. These two effects are reinforcing in the sense each increases the

possibility and extent of gains from trade from the perspective of the foreign country.

109

1 (ID,

Units of Truth

Foreign Indifference Curves

 

 

 

h. ’

Units of Beautv

Figure 4.4: Decision to Compromise: The Foreign Perspective

The case described above represents a unilateral compromise, however, a case
when the T ruth-standard counti'y lowers the standard by providing for greater flexibility
in allowing exceptions is not uncommon“). In this case both Truth- and Beauty-standard
countries are willing to adjust their standards. A small adjustment downwards from the
upper limit of a standard would leave the Truth-standard country at least as well off or

better off due to terms of trade effects, increased variety, and realized economies of scale.

Similar adjustment holds for the Beauty-standard country as well, leaving both countries

 

” Consider examples of trade with countries with different labor or environmental standards.

110

better off. Nevertheless, even if a potential Pareto improvement is possible, the political

difficulty in negotiating the access to the Pareto improving point is well known.

4.5.3. No Trade Occurring due to Large Differences in Standards

In this case standards are too far apart and there are no gains from trade to be had
— that is, there is no trade even if it is legally and technically possible (i.e. we do not
assume autarky). Countries endogenously choose not to purchase other goods, as there
are no benefits. Similarly, if economies of scale are not sufficient to justify harmonization
or compromise, no trade occurs. Graphically on Figure 4.3, this would occur when a
foreign indifference curve passing through A B on the horizontal axes would intersect the
vertical axes above I/ZTT, leaving foreign country better off by staying isolated rather

than trading and adjusting standards.

4.6. Two Countries Case with Equal Size Assumption Released

Now we allow the countries to differ in the number of workers available in the

economy keeping their per capita endowments the same. The Home country consists of

L” number of workers, Foreign of LF, where LH << LF . In this case economies of scale
anticipated by differently sized partners influence the decision to harmonize. In addition,
the number of varieties is an increasing function of labor available: the large country’s
number of varieties exceeds the number of varieties produced by the small country. For

any single variety the comparison is graphed in Figure 4.5.

111

Units of Truth

A ’Tq

    

IC large country

IC sm 1 country

Integrated PPF
A T

 

L

I; A '3' Units of Beauty

Figure 4.5: Small vs. Large Country

The small country’s autarky PPF (A TAB) lies below the large country PPF
(A ’7A ’3). In autarky Beauty prevails in the small country, while Truth prevails in the large
country. Combined open trade PPF is represented by a dotted line. Due to the differences
in size, the foreign country gains a small amount from trade with the domestic country.
The integrated PPF lies only slightly to the right of the large country PPF. Thus, the

potential gains for the foreign country are small, and the foreign country will be

112

unwilling to compromise on standards for these small gains. In contrast, access to trade

with a large foreign country can yield large gains from trade for the domestic economy.

4.7. Chapter Conclusions

We have developed a theoretical model of standard formation. In autarky,
national standards reflect the preferences of domestic consumers. The possibility of gains
from trade encourages countries to modify their standards to facilitate trade with other
countries having similar standards. If the gains from trade are sufficient, countries will
compromise or harmonize of standards to achieve these gains. In the case of countries of
similar siZe (bargaining power), compromise may be feasible. In the case of countries of
different size, harmonization of the smaller country's standard to that of the larger

country may be more likely.

113

CHAPTER 5

CONCLUSIONS AND BIOTECHNOLOGY APPLICATION"

The theory essays presented in the earlier chapters provide a well-suited basis for
analyzing one of the most challenging issues currently troubling trade relations between
the European Union and the United States: biotechnology and related trade in transgenic
crops and genetically engineered (GE) foods. In the context of countries’ preferences
discussed in this dissertation, the EU position lies close to the Truth boundary, while the
North American position lies close to the Beauty boundary.

The chapter abstracts from any legal analyses of the formal GE dispute filed in the
WTO by the US in August 2003, as well as any policy consequences such as labeling
requirements in the EU. Instead, it aims to apply the conclusions from the theory to a
pressing problem of divergent regulatory approaches, and discuss its implications. First,
this chapter presents a non-technical list of conclusions in a general form from the theory
chapters42 and a backgrounder on GE clarifying the differences between the EU and US.

Next, we apply the model to the GE case, and finally discuss its implications.

5.1. Conclusions from the Theory Essays

In autarky, national standards reflected the preferences of domestic consumers

regardless of the model considered. Chapter 2 utilized a model of endowed economies

 

" Earlier draft published as Tothova, M. and LP. Oehmke (2004) “Genetically Modified Food Standards as
Trade Barriers: Harmonization, Compromise, and Sub-Global Agreements”, Journal of Agricultural and
Food Industrial Organization: Vol. 2: No. 2, Art. 5 http://www.bepres_s.com/iafio/volZ/issZ/agfi

’2 The discussion of the social planner’s problem with different endowments (Appendix A2) is omitted.

114

forced to trade by the structure of the utility function: autarky utility was zero, showing
standards have the potential to be used for strategic reasons and influence the terms of

trade. Standards were modeled on a continuum of choices, allowing for a compromise.

Conclusions derived include, but are not limited to:

1. A Pareto optimizing decision by the social planner choosing standards
simultaneously for both countries, accounting for the extemality imposed on the
other country because of the difference between standards, and accounting for the
difference between relative endowments between countries. When countries were
of equal size and:

a. The difference between countries’ standards was small, social welfare was
maximized by allowing countries to “keep” their preference standards.
That is, two different standards were present in the global economy.

b. The difference between countries’ standards was large, social welfare was
maximized by introduction of a single standard in the middle of the
respective countries’ preference standards.

2. When countries forsake the terms of trade effect and considered domestic
consumers only, each country chose its choice parameter to be identical with its
exogenous preference parameters regardless countries’ sizes and differences in
endowments.

3. When countries considered the gains in welfare introduced by terms of trade
effects (technically they played a Nash game), the outcome depended on the
relative endowments of the countries. Unlike in the social planner problem, the

difference between standards did not play any role.

115

a. When the endowments across countries were identical, countries chose
their standards in the same Euclidian distance from their preference
parameters.

b. When the endowments across countries differed, the country with
abundant endowments exercised more power in standard selection, and
placed its standard closer to its preference parameter relative to the
location of the country with scarce endowments.

Chapter 3 considered a monopolistic competition with increasing internal returns
to scale, constant marginal costs, and love of variety bestowed on the consumers. Trade
in the model was induced by love of variety — however, trade was an option to achieve
higher welfare, rather than a requirement to achieve positive utility. Standards were
modeled as a choice between two discrete standards, with harmonization being a
unilateral decision. Model conclusions included, but were not limited to:

1. Trade without any harmonization (equivalent to consuming mix and match of
varieties, but never any variety in both specifications) was never welfare
decreasing, and almost always welfare increasing as compared to autarky.

2. Harmonization to a common standard had a potential to be welfare worsening —
when the “harmonization effect,” or the effect of a standard different from the one
preferred in autarky - exceeded the increased variety effect, and a country would
have vetoed the harmonization and trade.

3. When harmonization was found to be welfare improving, it necessarily brought
larger benefits the country whose standard was unilaterally adopted by the other,

harmonizing country. While both countries benefited from the larger number of

116

varieties available, the harmonizing country discounted the consumption in the
standard that suboptimally matched its preferences.

Extending to a larger number of countries, when countries found harmonization to
a common standard welfare worsening, they may have vetoed the formation of a
global agreement. Instead, they clustered in two (same as number of standards
considered) sub-global agreements composed of like members to avoid problems
with global harmonization, although the gains from trade could have been smaller
in the sub-global harmonization than in a global agreement.

Chapter 4 kept some, while released other, assumptions from Chapter 3. As in

Chapter 3, trade was not a requirement, but an option. Trade was mostly driven by

economies of scale achieved by producing on a common standard, and to a lesser impact

by introduction of additional varieties. As in Chapter 2, standards were modeled as a

continuum of choices, and compromise was allowed. Thus,

1.

Harmonization (understood as one country altering its behavior) to a common
standard occurred only when the economies of scale were sufficient enough to
yield higher utility, including compensation for the consumption of a less
preferred standard.

Compromise (understood as both countries altering their behavior and agreeing to
a common standard different from their respective autarky standards) only
occurred when the economies of scale were sufficient to justify the change, and
both countries were compensated by higher utility from consuming larger amount

of goods and/or more varieties at a less-preferred standard.

117

3. If standards were too far apart and / or economies of scale were not sufficient to
justify a change in standards, gains from trade were insufficient to justify
harmonizing or compromising standards.

4. Due to larger benefits, smaller country as more likely to adjust its standard to the
larger country standard.

5. When harmonization or compromise were not feasible options (economies of
scale not sufficient for compensation), remarks on sub-global agreements from

Chapter 3 apply.

5.2. Backgrounder on Genetic Modification

Genetic modification (also “gene splicing”, “recombinant DNA technology” or
“genetic engineering (GE)”) is defined as the manipulation of a living organism's genetic
make-up by eliminating, modifying or adding copies of specific genes, often from other
organisms [and from other species] through modern molecular biology techniques."3
Biotechnology is the application of GE to living organisms to develop new products and
processes."4 Two applications of genetic engineering exist: transgenic and non-
transgenic. Transgenic modification adds new material from different biological
organisms, and hence, could be considered a new product characteristic under the WTO.

Non-transgenic modifications simply select genetic material from the same species

(Phillips and Kerr, 2000).

 

’3 http://www.iic.bbsrc.ac.uk/exhibitions/bio-future/glossarv.htm. viewed June 20, 2004
“ Id. For a more scientific treatise of GM and biotechnology refer to Bemauer (2003), brief overview in
Runge and Jackson (2000).

118

Some of the often repeated terms accompanying GE crops and foods are identity
preservation, segregation, and traceability. Identity preservation is a system of crop or
raw material management which preserves the identity of the source or nature of
materials. Segregation implies setting up and monitoring separate production and
marketing channels for GE and non-GE products. Traceability measures covering feed,
food and their ingredients includes the obligation for businesses to ensure that adequate
procedures are in place to withdraw feed and food from the market where a risk to the
health of the consumer is discovered.45 Lacking mechanisms for identity preservation,
segregation and traceability are likely to cause the entire production of a certain crop of

which some is GE, to be considered “contaminated” with the GE kind.

5.2.]. Economic Importance

The first GE products were approved for commercialization in the United States
in the early 19905. According to the ISAAA (2004), in 2003 among the major biotech
growers (all crops) are the United States (accounting for 63 percent of global total),
Argentina (21 percent), Canada (six percent), Brazil (four percent), China (four percent),
and South Africa (one percent). Small areas were planted in Australia, India, Uruguay,
Romania, Spain, Germany (limited area), Bulgaria, Philippines, Indonesia, Columbia,
Honduras, and Mexico (ISAAA, 2004). The most widespread crops are com, soybeans,
canola, and cotton (ISAAA, 2004). Among other crops are carnation, chicory, flax,

melon, papaya, potato, rice, squash, sugar beet, sunflower, tobacco, tomato, and wheat“.

 

’5 Definitions of identity preservation, segregation, and traceability are from
ht_tp://europa.eu.int/comm/agriculture/publi/gmo/glossary.h_tm, viewed on June 20, 2004.
’6 hgp://www.agbios.com, accessed on June 23, 2004.

119

About 99 percent of the crops involved two key input traits - herbicide tolerance and
insect resistance — while less than 1 percent of the crops had other input or output traits
(e.g. improved nutrition) (Phillips and Kerr, 2000). In 2003, herbicide tolerance,
deployed in soybean, maize, canola and cotton occupied 73% or 49.7 million hectares of
the global GE 67.7 million hectares, with 12.2 million hectares (18%) planted to Bt crops.
Stacked genes for herbicide tolerance and insect resistance deployed in both cotton and
maize continued to grow and occupied 8% or 5.8 million hectares (James, 2003).

In 2003, the global market value of GE crops was estimated to be $4.50 to $4.75
billion, having increased from $4.0 billion in 2002 when it represented 15% of the $31
billion global crop protection market and 13% of the $30 billion global commercial seed
market. The market value of the global transgenic crop market is based on the sale price
of transgenic seed plus any technology fees that apply. The global value of the GE crop

market is projected at $5 billion or more, for 2005 (James, 2003).

5.2.2. Reasons for Controversy and Technology Regulation

The controversy over GE food has spread from initial concerns about consumer
health to include concerns over environmental risks, world food security, corporate
dominance of the food supply, ethical issues, farm-level economics, and regulatory
problems (Bemauer, 2003). Sheldon (2003) divides concerns concerning GE ingredients
in food products into a series of ethical concerns over the science of biotechnology (not
accepting genetic engineering as an extension of traditional breeding), concerns relating
to food safety (i.e., introduction of new allergens by using a protein from nuts, resistance

to antibiotics), and the impact of GE crops on the environment (pollen drift, modified

120

resistance of naturally occurring counterparts) (italics in the original). Attitudes towards
“white” biotechnology (medical and environmental) are very positive across the globe,
while attitudes toward “green” biotechnology (field crops)“ differ (e. g. Sheldon (2003),
Bemauer (2003), etc.). The public may be unaware of auxiliary substances used in
processed food, like GE enzymes used in cheese and beer production.

In principle the nature of the GE is similar to the problem of processes: as of now
the WTO does not recognize a production process as a product’s attribute if the product
made using process A is indistinguishable from product made using process B.
Mandatory standards may be applied domestically that, for example, limit the use of
energy in production process, but an imported product cannot be discriminated against
only because the process was energy-intensive; equivalently, imported products that are
physically the same as domestically produced ones are considered to be “like” domestic
products irrespective of how they were produced (Sampson, 2000). In the same manner
one cannot legally differentiate GE corn from its natural (wild) or conventional
(traditional breeding) counterparts.

Any advanced technology —including biotechnology — characterized by a large
information gap between the producers of the innovation and the intended consumers
brings up new regulatory issues (Isaac and Kerr, 2003)“, and different perceptions of
“risk”. Risk is defined as a scientific determination of the relationship between cause and
effects in situations where adverse effects can occur (Samson, 2002). Risk assessment is

defined as the evaluation of the potential for adverse effects on human or animal health

 

‘7 Some sources categorize biotechnology into red (medical), green (agricultural), and white (industrial —
i.e., making advanced enzymes, biological manufacture of plastics and fuel) (survey of biotechnology in

Economist, March 27, 2003)
‘8 Please see Isaac and Kerr (2003) for a discussion on technology regulation in the Risk Analyses

Framework.
1 2 1

arising from the presence of additives, contaminants, toxins or disease-causing organisms
in food, beverages or feedstuffs (Pauwelyn, 1999). Risk management is a determination
of the acceptable level of risk and the selection of [SPS] measure or other necessary
standards to permit risk to be managed appropriately to meet that level (e.g., Pauwelyn
(1999); Samson (2002)).

The precautionary principle is based on the notion that regulation should prevent
damage occurring from a particular action rather than letting it arise and then dealing
with the consequences (Perdikis, 2000). It states that where there are threats of serious or
irreversible damage, lack of full scientific certainty shall not be used as a reason for
postponing cost-effective measures to prevent the eventual outcome of risk, covering the
gap between banning a product or procedure until science has proved it is harmless and
not banning it until science has proved that there is a real risk (Sampson, 2000). McNelis
(2000) describes it as a cautious approach to managing potential threats to the
environment, human, animal, or plant health.

Substantial equivalence is often cited as an antonym of the “precautionary
principle,” and as a synonym of “scientific approach.” A concise definition of substantial
equivalence is hard to achieve: one of the few identifies it as a concept that embodies the
idea that existing organisms used as foods, or as a source of food, can be used as the basis
for comparison when assessing the safety of human consumption of a food or food

component that has been modified or is new (OECD, 1993 in Millstone et al, 1999).

122

5.3. Controversy: EU vs. US

Because the US and EU are major trading partners, unresolved regulatory issues
have negative effect on mutual trade relations. For example, after the introduction of GE
corn to the US, corn exports to the EU fell by 95% due to refusal of the EU to accept
shipment which contained a small amount of GE corn.

The first GE product was exported by the US to EU in 1996 — tomato puree from
Califomia was voluntarily labeled as genetically engineered and, due to its price
advantage compared to its conventional counterpart, it became highly popular with the
consumers (i.e., Carter and Cruere, 2003). However, imports of GE soybeans to Britain
later that year draw an opposition from environmental groups such as Greenpeace49
(Carter and Cruere, 2003; Bemauer, 2003). Later, the countries adopted different
approaches to GE regulations50 (US endorses the substantial equivalence, the EU
precautionary principle), resulting in the current biotech/GE puzzle of differing
regulations impeding trade resting on the divergence in societal choice, science,
principles and standards among countries.

Consumers’ attitudes with respect to GE are often categorized with the US being
pro-biotech, and the EU being against. Runge et al (2001) provides an overview of the
historical and cultural factors that have contributed to divergent US. and European views

on GMOs, and to resulting different national regulatory approaches for these products,

specifically labeling policy. Sheldon (2003) offers a comprehensive discussion of the EU

 

’9 Greenpeace consists of Greenpeace International (a Council based in Amsterdam), and offices around the
world. Greenpeace International, among others, monitors organizational development of local offices,
coordinates global campaigns, monitors compliance with core policies, etc. As of 2004, “Say No to Genetic
Engineering” is one it its global campaigns (http://wwwgreemreaceorg, accessed June 28, 2004).

5° For a book-length review on regulation polarization please refer to Bemauer (2003).

123

legislation related to GMOs. The key pieces of EU legislation include two directives
adopted in 1990 concerning the management of GMO research and development and the
deliberate release of GMOs), the Novel Food Regulation (NFR) adopted in 1997
(establishing an approval procedure for novel foods and novel foods ingredients defined
either as foods or food ingredients containing or consisting of GMOs, or foods and food
ingredients produced from but not containing GMOs), and the formalized moratorium on
GMO approval in form of an amendment to a directive from in 1999 (lifted in April
2004). The last directive also updated the labeling rules: all food containing ingredients
that have been derived from GE crops has to be labeled, irrespective of whether the
relevant rDNA or proteins are still detectable, although accidental traces up to 0.9 percent
of authorized GE material is exempt from labeling (Sheldon, 2003). The main tensions
are caused by the (former) moratorium and the labeling requirement. The EU regulatory
approach focuses on “process” as well as “product” in an effort to handle consumer
concerns and demands for certainty of the safety of the new biotechnologically modified
products. Risk-averse regulators have tended to pursue certainty, with the result that
product approval has generally been withheld until all risks have been eliminated
(Buckingham and Phillips, 2001).

In the US, biotechnology is regulated through three existing agencies: USDA’s
Animal and Plant Health Inspection Service (APHIS) — regulating permits for or
providing notification of the introduction of a GMO into the US, the Environmental
Protection Agency (EPA) — responsible for regulating plants that are GM to express
pesticides, and finally the US Food and Drug Adrrrinistration (FDA) - dealing with the

pre-market approval of GMOs and foods containing GM ingredients, and also providing

124

guidelines on the labeling of GM foods (Sheldon, 2003). The US, unlike the EU,
advocates the policy of substantial equivalence, recognizing genetic engineering as an
extension of traditional breeding, and requiring labeling of GE if and only if the GE
product is substantially different from its conventional counterpart (i.e., presence of
allergens, advanced nutrition, etc.). In all other cases, a GE product is considered as safe
as its conventional counterpart. The regulatory approach taken by the US is consistent
with recommendations for assuring the safety of GMOs made by the OECD and Codex
(Sheldon, 2003). The North American model used by Canada and the United States
focuses on products rather than processes and allows for the use of existing agencies and
risk assessment methodologies (Buckingham and Phillips, 2001).

The case can be made that the EU standards -— resulting from its regulatory
approach — act as a barrier to trade, but a barrier that in a relatively unchanged form will
withstand challenges made via the WTO. The barrier does not arise because of border
measures but because of differences in domestic regulatory approaches (Isaac and Kerr,
2003). The conflict of underlying principles behind polarization of regulatory
frameworks has fundamental consequences for the WTO, multilateral environmental

agreements (MEAs), and the relationship between trade and environment.

5.4. Model Application

For practical purposes this chapter assumes the EU and US are equally endowed
countries. While causality is open to discussion, we argue that different regulatory

approaches in fact reveal country’s preferences for transgenetic crops and GE foods. In

125

the context of the models, we arbitrarily select that the EU, applying precautionary
principle, reveals preference for Truth (zero tolerance), the US, endorsing substantial
equivalence, discloses preference for Beauty (full acceptance). In autarky, the EU and US
standards reflected the preferences of domestic consumers.

Chapter 2 worked with a model where trade was a necessity: countries’ utility was
zero unless trade took place. The social planner’s decision was dependent on the distance
between the standards. Since it is impossible to quantify the difference between the
“precautionary principle” and “substantial equivalence,” we consider both cases. If
indeed the difference between the EU and US was “small,” it is socially optimal to let the
countries trade at their respective preference parameters: the EU applying precautionary
principle, the US applying substantial equivalence. However, as the EU is not willing to
accept GE products, it is more probable that the difference between the “precautionary
principle” and “substantial equivalence” is large. In that case the social planner would
“split the difference.” Anecdotal evidence on the EU imports of soybeans seemingly
supports the argument: as most of the soybeans produced in the US are GE, the EU
imports soybeans from Brazil. Brazil law does not allow production of GE; another
anecdote claims cross-trade in GM between Brazil and Argentina, and lack of intellectual
property protection. The EU might not have the necessary regulation to get “half way,”
however; imports from Brazil could practically be positioned in the middle of the interval
to account for the terms of trade extemality. Returning to the EU — US case, if the EU
and US forsake the terms of trade effect and considered domestic consumers only, each
country would choose its choice parameter to be identical with its exogenous preference

parameters (zero tolerance and full acceptance). Conclusions of Chapter 2 demonstrated

126

that under certain circumstances the social welfare maximizing outcome is emergence of
two different standards: which, translated into the GE case, implies the social welfare is
maximized by applying the precautionary principle in the EU, and substantial
equivalence in the US.

Chapter 3 showed that non-harmonized trade between the US and EU (and,
consequently, around the world) would be always welfare improving - assuming absence
of any trade restricting policies. The EU and US policies regarding GE impose a certain
standard (or perhaps lack of it in the US), and thus have a potential to be welfare
worsening. The EU valuation of the GE products is very low, positioning it to the very
left (possibly zero) on Figure 3.4. The US valuation is very high, positioning it to the
very left. Any gains in the number of varieties if harmonization took place would be
offset by the consumption on a different standard. Both countries would likely veto
harmonization and trade. Adding other trading partners to the picture (to be discussed
momentarily on the evidence from developing countries) results in formation of sub-
global agreements, each using a different regulatory approach.

Chapter 4 explored harmonization and compromise in the traditional economies
of scale argument. It is arguable whether there are economies of scale in agriculture, and
it was already stated the differences in standards between the EU and US are high. Thus,
the economies of scale are not sufficient to justify harmonization or compromise of
standards. Consequently, earlier remarks on the emergence of sub-global agreements
apply. Conclusions from Chapter 3 and 4 on formations of sub-global agreements in case

harmonization of standards is required are confirmed in the GE case.

127

5 . 5. Anecdotal Evidence from Third Countries: Small vs. Large

The transatlantic GE “conflict” does not remain confined to only two countries.
Rather, third countries wishing to conduct business with either of the leaders (EU and
US) ought to observe their protocols, and are often caught in between, potentially with
insufficient resources to conduct scientific assessments, and possibly with safety of GE
not given priority on their agenda. In these cases the decision is often based on trying to
secure potential markets for their agricultural exports. The dominant strategy might be to

not adopt GE varieties (albeit their adoption might produce higher and less labor
intensive yields) and thus potentially have access to both types of markets.

The transatlantic regulation polarization influencing the rest of the world can be
seen on the example of China. China appears to be primarily concerned with producing
food and fiber for internal needs. It leads the world in public biotech crop research
(Huang et al, 2002 in Carter and Gruere, 2003), it is conducting transgenic research on
over 130 species, and in 2002 transgenic crop area exceeded 2 million hectares (Gao,
2003). GE cotton accounts for about 30% of China’s cotton acreage. In the field trial
stage are GE rice, wheat, soybeans, potatoes, cabbage, and tobacco. While China has yet
to announce a firm position on GE labeling, it has recently proposed restrictions on GE
crop imports. Outside China, this is viewed as a trade barrier that limits soybean imports
from the United States. Despite its labeling requirements, it has recently approved five
transgenic US varieties (one corn, two soybean and two cotton) for import (ISAAA,
2004). China’s position towards biotechnology in agriculture appears to be heavily
influenced by EU policy (Carter and Gruere, 2003); while at the same time it has widely

adopted GE research and crop production. India appears to be following suit in adopting

128

biotechnology production, albeit at a slower pace. These are populous countries and even
though they may be low-cost producers on a per unit basis, the large populations mean
that these countries face high costs of feeding their populations. At a national level the
aggregate cost reductions from GM production may be very significant (Tothova and
Oehmke, 2004).

In contrast, some African countries have apparently given higher priority to
potential export markets than to existing domestic food needs. Malawi, Zambia and
Zimbabwe all faced severe food shortages due to drought in 2002. The US. offered food
aid to these countries, much of which was in the form of GE corn. Even though each of
these countries has been a net importer of com/maize over the past several years, they all
balked at receiving GE food aid. Eventually Malawi and Zimbabwe accepted the GE corn
if it was previously milled or milled immediately upon entry to the country (to prevent
gene escape with possible damage to trade); Zambia never did accept the aid.

Namibia has an EU import quota that allows it to export 13,000 tons of beef to the
EU (Republic of Namibia). In order to feed the cattle, Namibia was set to import maize
from South Africa. However, South Africa allows the use of GE maize, which is widely
adopted by farmers. Because Namibia (nor South Africa) could guarantee that the
imported maize would have been GB free, the EU said it would refuse to accept
Namibian beef that had been fed South African maize. Namibia found alternative sources
for GE-free cattle feed (Fox, 2003).

Egypt actively developed a GE local (brown) potato variety (Spunta) with locally
valuable pest resistance characteristics (tuber moth resistance). Egypt sought US

collaboration in the genetic modification, and established regulatory infrastructure,

129

intellectual property protection and biosafety protocols (Madkour et al, 2000). However,
fear of gene escape to export (white) potato varieties and consequent loss of European
export markets was sufficient to quash commercialization of the GE variety.
Interestingly, some countries have defacto accepted GE agriculture even though
they don’t quite have the necessary infrastructure in place. For example, Argentina has
the second largest area planted to GE crops, even though there is anecdotal evidence that
local companies are selling herbicide tolerant varieties without appropriate royalty
payments (increasing herbicide sales could explain some of the patent owner insouciance
toward this situation) (i.e., Qaim and de Janvry, 2003). In Brazil, there is increasing
evidence that herbicide tolerant soybeans are being adopted, despite an official stance

disallowing GE crops.5 ’

5.6. Deeper Conflict: Countries vs. Multilateral Agreements

The three sets of multilateral treaties most relevant (to standards) for GMOs and
foodstuffs are the Cartagena Protocol on Biosafety (CPB), in principle an environmental
treaty; the Trade-Related Aspects of International Property Rights (TRIPS), negotiated as
part of the Uruguay round of the General Agreement on Tariffs and Trade (GATT); and

the Codex Alimentarius Commission (Codex) allied directly with the Agreement on the

 

5’ “Illegal GMO soybeans give Brazil ‘unfair advantage’”, Delta Farm Press, Sept. 19, 2003. Available at
http://www.findarticles.com/p/articles/mi mOHEE/is 37 60/ai 108087663 last viewed on June 23, 2004.

130

 

Application of Sanitary and Phytosanitary Measures (SP8) and in some way indirectly
with the Agreement on Technical Barriers to Trade (TBT)”.

The CPB ensures that the transfer, handling, and use of living modified
organisms53 (LMO) (italics added) does not have an adverse effect on biological diversity
and impose risk on human health (Sheldon, 2003). Its key element is the Advanced
Informed Procedure, i.e., a prior notification and consent procedure for the export and
import of LMOs (Eggers and Mackenzie, 2000). The CPB does not establish any
standards or labeling requirements regarding processed GE foods (pharmaceuticals are
excluded completely). It establishes some minimum rules regulating the exportation of
transgenetic seeds for field trials or commercial cultivation. The regulation of GE foods
and feedstuffs, which make up 90 percent of international trade in GE products, is largely
lefl to the parties. Similarly, detailed identification requirements only apply to LMOs that
are to be introduced into the environment, while shipments of LMOs intended for feed or
food or processing have to be labeled as “may contain LMOs.”5" The CPB does not
directly address the contentious issue of domestic labeling, and it does not determine how
to deal with the scientific uncertainties arising from GMOs or give any clear directives
when a Party should prohibit or restrict the import of a particular LMO (Eggers and
Mackenzie, 2000). The EU and the US disagree on the precautionary principle that is
endorsed by the CPB (the US. is not a signatory to CPB) (Kerr, 1999; Caswell, 2000;

Eggers and Mackenzie, 2000; Phillips and Kerr, 2000, Maredia et al., 2001).

 

’2 Also applicable is the work of the International Office of Epizootics for animal and International Plant
Protection Convention for plant health (Buckingham and Phillips, 2001). Standard developed by these
organizations are not binding and serve as reference points.

53 The definition of LMOs was part of the negotiation: the Miami Group (comprising the US, Canada,
Australia, Argentina, Chile and Uruguay) argued that lengthy approval procedures are not justified for
commodities which are not intended to be released into the environment, and hence would not impact the
conservation and sustainable use of biological diversity (Eggers and Mackenzie, 2000).

5’ Full text of the CPB is available at http://www.biodiv.or2/biosafetv/protocol&p, accessed June 23, 2004.

131

North America and the EU are basically in accord on intellectual property rights,
and the TRIPS agreement is not a major source of friction between the two (Tothova and
Oehmke, 2003).

The Codex Alimentarius was created in 1963 by the Food and Agriculture
Organization and the World Health Organization to protect health of the consumers,
ensure fair trade practices in the food trade, and promote coordination of all food
standards work undertaken by international governmental and non-govemmental
organization (Codex Alimentarius, 2003). Although the Codex was not originally given
direct powers to set binding standards, it is explicitly recognized under the WTO SPS
Agreement as setting food safety standards, guidelines and recommendations. Codex is
currently developing principles for the human health risk analysis of GM foods with non-
binding effects on national legislations. Codex standards for GM foods presumably will
not be forthcoming until afier the risk assessments are completed.

The SPS and TBT Agreements were discussed in Chapter 1. The SPS Agreement
itself dictates that measures to be taken to ensure food safety and animal and plant health
should be based in the analysis and assessment of objective and accurate scientific data
(Sampson, 2000), but it also explicitly recognizes members’ sovereign right to adopt
higher standards that may restrict international trade, assuming the country provides
scientific justification of such a measure (WTO, 2004).

Different regulatory policies, presented as consequences of underlying regulatory
approaches applied in the US and EU — and at the same time, foundations of a trade

conflict — are indicative of multilateral international treaties: Codex, for example,

132

endorses the scientific evidence approach, while CPB gives its backing to the
precautionary principle.

The WTO has the ability to address issues raised under the umbrella of GATT
Article I (non-discrimination), 111 (national treatment), and XX (country’s right to address
policies protecting human, animal and plant health), as well as SPS and TBT agreements
relying on the concept of like goods, and it provides a natural outlet to channel trade
related disputes (Isaac and Kerr, 2003). In the GE dispute the WTO is challenged to
determine an issue of domestic regulatory competence, namely which regulatory
approach — scientific rationality and substantial equivalence or social rationality and the
precautionary principle — is consistent with trade rules and when these regulations are
appropriate from a trade perspective (Isaac and Kerr, 2003).

The WTO does not appear to be properly equipped to entertain disputes
interrogating the precautionary principle, and thus indirectly questioning the legitimacy
of the CPB (i.e., Isaac and Kerr, 2003). Whether or not the EU “embargo” is in fact a
trade barrier depends on what paradigm -— substantial equivalence or the precautionary
principle — would be adopted by the panel. If the US position of substantial equivalence
was adopted, then the EU position would imply a discriminatory, less favorable
treatment, and thus would be considered a trade barrier. However, the panel is not in the
position to condemn domestic regulatory policies, especially when in compliance with
the CPB. Nevertheless, the presence of the CPB brings up yet another issue. The CPB is
55

in fact an environmental treaty, existing independently of the WTO and its signatories.

As a matter of fact, the US has not became a signatory to the Protocol — leaving it in a

 

’5 Detailed discussion on the relationship between the Cartagena Protocol and WTO is in Eggers and
Mackenzie (2000).

133

position to be completely aside, and not bonded by its clauses, and putting potential WTO
panel wishing to consider the Protocol in a unwelcoming position of reconciling
(subglobal) CPB (or other environmental agreements) with a (global) GATT, SPS and
TBT Agreements (e. g. Sheldon, 2003; Eggers and Mackenzie, 2000)

The CPB may be reasonably well designed to deal with issues related to trade in
GMOs that will enter agronomic or aquaculture production, but it seems poorly designed
for regulating trade in GE products (Phillips and Kerr, 2000). As such, the CPB, lacking
clear linkages with the WTO, is not an amicable settlement between the US and EC on
their trade conflicts regarding the GMOs (Eggers and Mackenzie, 2000). The position of
the WTO, eager to maintain the divide between domestic and trade policies though the
rigid application of four principles (focus on products and not PPMs, national treatment,
most favored nation, and common exemption of environmental and natural resource
issues under GATT 1994), is likely to be challenged by agricultural biotechnology
centered around PPMs (Phillips and Kerr, 2000). The CPB, although aimed to coexist
with the WTO, re-opened the doors to the trade and environment discussions by allowing
considering process rather than final product characteristics. Isaac and Kerr (2003) claim
the CPB is the only one stirring such an attention from a large pool of multilateral
environmental agreements and challenging the WTO, mostly because other agreements
(such as Basel Convention, Montreal Protocol, etc), are (1) very specific, pertaining only
to the transboundary movement of very select products and not to broad arrays of
products with quite disparate end uses such as GMO, and (2) in agreement across

Atlantic, meaning they are represented by similar regulatory approaches, and there are no

134

significant differences in the systemic principles and frameworks that provide the

foundation for regulation.

5.7. Implications for the World Trading System

Bemauer (2003) concludes that prevailing public and private sector policies in the
US and EU do not add up to an effective strategy for mitigating or overcoming regulatory
polarization.56 Since trade policies originate in domestic regulatory frameworks, without
the harmonization of the domestic regulatory frameworks the harmonization of the trade
policies is unlikely.57

If the EU wins the GM dispute and that the panel endorses the EU regulatory
approach by making a decision against the science-based, rules-based trading principles
at the heart of the WTO, and legitimizes the use of discretionary, protectionist measures
(Isaac and Kerr, 2003). In this case the WTO essentially endorses different domestic
regulatory frameworks and trade will not occur. On the other hand, if North America
wins, a WTO decision will be portrayed as both a positive decision for biotechnology and
a negative decision against human, animal and environmental health and safety
regulations in the EU as well as a decision against the CPB and the protection of
biodiversity (Isaac and Kerr, 2003). The criticism of environmentalists that trade

liberalization is only achieved at the expense of environment is inevitable.

 

’6 Isaac and Kerr (2003) use term “transatlantic regulatory regionalism”.

’7 Not to blame the US and EU for not being willing to cooperate: according to Buckingham and Phillips
(2001) agreed to the Trans-Atlantic Economic Partnership, established under the New Trans-Atlantic
Agenda in 1995 to “tackle regulatory barriers and to improve regulatory and scientific cooperation in the
fields of consumer and plant health, biotechnology, and environment’. Similar partnership is in place
between the EU and Canada. Also in May 2000 the US and Europe announced they would create a bilateral
high-level discussion group, which would look at the risks and benefits of GM crops and foods
(Buckingham and Phillips, 2001). The GM issue also regularly shows up in the G-8 meetings.

135

As a result of a current lack of consensus between the EU and US as world
leaders, a global consent on the GM issue (allowing for non-harmonized trade via mutual
recognition, or endorsing a common approach) is likely to be a matter for the distant
future. At the moment the polarization in regulatory approaches applied by the trade
leaders is being transposed into the polarization of the trading partners in accordance to
the regulatory principle they choose — and their implementation.

Consequent to the polarization of regulatory approaches would be a split of the
trading partners into two parallel “alliances” — the “precautionary principle alliance,” and
the “substantial equivalence alliance”. The former is likely to be slow to approve new
GM varieties (especially green biotechnology), and would tend to require mandatory
labeling regardless of whether the product in question would be regarded as safe. The
latter alliance, in contrast, would be more liberal in approving GM varieties and would
require (mandatory) labeling if and only if the products would be different from their
conventional counterparts nutritionally or otherwise. The alliances -— smaller, more
homogeneous groupings of countries — would be facilitated by their respective leaders:
the EU in the “GM-fiee/precautionary principle alliance” and the US in the
“GM/substantial equivalence alliance.” The CPB allows parties to proceed according to
their own domestic regulatory framework, adopt simplified procedures, or enter into
bilateral or regional agreements as long as these are consistent with the objective of the
Protocol (Eggers and Mackenzie, 2000).

This notion is consistent with the findings of the theory essays, and the evidence
on increasing numbers of small trade agreements of all types (not necessarily related to

biotechnology), and fully supported by social choice and club theory (i.e., Buchanan,

136

1965; Olson and Zeckhauser, 1966; Olson, 1968; Comes and Sandler, 1996). “Alliances”
or “clubs” are formed by a relatively homogeneous set of countries where homogeneity
can be defined in terms of similar consumers’ attitudes, industry structure, interest
groups, income, tolerance levels, similar perceptions of risk, etc. Groupings of countries
with similar regulatory approaches in the biotech case become defacto clubs in a
Buchanan sense, where standardized goods (or non-standardized goods, depending
whether precautionary principle or scientific equivalence was adopted) have the
characteristics of club goods, and non-members are excluded from consuming the
benefits — that is, unrestrained trade is conducted among members of the club. Rege
(2002) claims that harmonization of policy goals is more feasible on a “regional” basis
(quotation marks added), particularly in the case of economic groupings that have made
considerable progress in economic integration, than on a multilateral basis particularly. A
club, for example, can develop common standards}8 agree on certain procedures
identifying previously negotiated requirements, or require traders to label their product (if
applicable). Such steps could be perceived as a tax or a technical barrier to trade by non-
members.

Unavoidably, the polarized “club” outcome is accompanied by a number of
dilemmas. Although the argument of securing welfare maximizing gains from trade is a
valid one, it comes at the expense of a separating equilibrium — and lower welfare than
non-harmonized trade. The choice of a club is also determined by the positive

extemalities for smaller countries of belonging to a certain club, such as economic

 

5’ For example, Australia and New Zealand cooperate in developing GM standards, as are the Canadian
Food Inspection Agency and the USDA Animal and Plan Health Inspection Service “which seeks to study,
compare and harmonize where possible the molecular genetic characterization components of the
regulatory review process for transgenetic plants” (Buckingham and Phillips, 2001).

137

networks, political allies, etc. For a small country that has less to offer a large country (or
club) in terms of gains from trade, there is a greater incentive to harmonize its domestic
standards with the large-country standards, even if these standards are not what the
country would have chosen without the trade considerations (e. g., the Zambia food aid
case). In addition, the separation might reflect neither consumer preferences in respective
countries nor the chosen regulatory approach to transgenic crops and GM foods. Rather,
it might be representative of trade and the “political” affiliation of countries. This
hypothesis remains to be empirically tested — assuming a data set allowing such analysis
would be available. As of now, GE and GE-fiee transgenic crops are treated as like
products, and their major producers (US, Argentina, Brazil) do not have separate
distributions channels to market their crops (with the exception of the organic channel
and a few contract farming situations). Therefore, unless proven GE-fi’ee, all products are
considered to be GE.

Isaac and Kerr (2003) see similar fragmentation of international markets as an
obstacle in recouping research and development costs sunk into product development —
countries would be forced to choose between a North American/WTO-style approach or
an EU/CPB-style approach to regulating GMOs, depending upon which side of the
Atlantic is deemed the more important market. Countries wishing to export their own
products to the EU will welcome neither GMO products nor GMO technologies, and thus
reducing the scale benefits arising from international market access.

The legal status of sub-global alliances is also questionable. The “clubs” might
not be institutionalized and the WTO notified in the spirit of the GATT, and in principle

function as a dejure agreement. As almost all countries belong to the WTO, disputes

138

similar to the current GM dispute are likely to arise again since the underlying regulatory
preferences will de facto be perceived as a trade barrier — as it is now. “Clubs” might also
choose to notify the WTO and function as formal trade agreements”. However, in that
case, the WTO places three relevant restrictions on trade agreements: the agreements
should not help nations in the agreement to the disadvantage of those not in the
agreement; domestic products and imports should be treated alike; and standards may be
set only on products, not on the production processes (WTO, 2004). In addition, with a
web of overlapping trade agreements already in existence, the effect of formalized trade
agreements based on a common regulatory approach remains to be seen. Alliance of the
countries based on their regulatory preferences is likely to reopen the issue of trade and
environment, resulting in sub-global environmental clubs.

Still, some countries might not fit the profile (precautionary principle, labeling vs.
substantial equivalence, no labeling) suggested here. For example, Australia and New
Zealand each employ the substantial equivalence approach, while requiring mandatory
labeling justified by consumers’ right to know in order to make an informed decision. It
is likely the “hybrid” countries would find it easier to penetrate into the “precautionary
principle club” market, since having a mandatory labeling experience would be

considered an advantage.

 

’9 Chapter 1 demonstrated how countries are allowed to formed sub-global agreements in the WTO
framework.

139

5.8. Chapter Conclusions

While both the EU and US would benefit from harmonization or a compromise on
standards in the biotechnology arena, the latter case study shows that a position uniting
the EU and US is unlikely due to the systemic principles underlying the regulation of
GMOs. In addition, the EU’s socially rational approach is consistent with the CPB and its
focus on PPMs, the scientifically rational North American regulatory approach is
consistent with the WTO and its focus on end-use products (Isaac and Kerr, 2003) — with
the US arguing that the EU’s precautionary principle acts as a barrier to trade.

Consequent to the polarization of regulatory approaches we notice polarization to
trade partners. Currently, two “trading” blocs are forming. The first consists of those
countries with restrictive standards regarding the commercialization and sale of GE crops
and foods. This group consists largely of the EU and their smaller trading partners,
although Japan and some other countries have similar standards. The second group
consists of those countries that have found GE crops to be essentially the same as or
“substantially equivalent” in US regulatory language - their non-GE counterparts. Each
of these trade blocs can make legitimate arguments that their stance is in the best interest
of their citizens.

The losers in this bifurcation of countries appear to be the developing countries,
most of whom are agriculturally based. These countries face a tension between lower
production costs by adopting GE crops and the maintenance of export markets in
conventionally grown varieties. In some food-insecure cases, the tension is severe enough
to force a tradeoff between export markets and food availability. While the US, EU, and

other affluent countries were framing their regulatory policies based on domestic

140

situation and preferences, some of the smaller agricultural export-oriented countries,
might find themselves choosing biotech regulatory frameworks to fit their trading
partners.‘50

Based on Isaac and Kerr (2003), the transatlantic GE crisis sends out at least two
signals (if North America win the current dispute). First, the EU as a pillar of
international trading regime is willing to ignore its WTO obligations. Second, when a
member does not agree with a WTO ruling, they can simply revert to a regional (in the
language of this chapter, sub-global) approach among like-minded countries aimed at
protecting them from the reach of the WTO. Regionalism could emerge as a protectionist
stumbling block to multilateralism rather than a constructive stepping-stone.

GE regulations and disputes in fact open the door to other issues, such as labor,
environmental, animal welfare regulations and standards on the WTO level, potentially
questioning national sovereignty granted in Article XX of the GATT. Indeed, Runge and
Jackson (2000) see GMOs as an issue resonating well beyond agriculture, helping to
mobilize and align consumer, environmental and health advocates in opposition to further
trade liberalization. With GMOs indirectly challenging the CBP, rather than encouraging
the coordination and cooperation between trade agreements and MEAs, it would justify
MEAs as a countervailing force to the environmentally insensitive WTO (Isaac and Kerr,
2003).

Some blending of regulatory principles is inevitable too: while the US claims to

Operate on the “sound science” based substantial equivalence principle, it too is

 

6° At the meeting in February 2004, developing countries insisted on tighter rules for trade in bioengineered
crops, prompting dismay in the US and other exporting nations...
ht_tp://www.foodtraceabilitvreport.com/eioumals/articles/demo article.§p?id=80592. accessed June 24,
2004

141

beginning to incorporate traits usually associated with precautionary principle such as
traceability and identity preservation, especially in the view of the Starlink episode61 and
as a means of retaining export markets.

Biotechnology has evolved as a complex matrix uniting issues that touch on
scientific, political, social, ethical and economic concerns, processes regulation, food
safety, environment, human, animal, and plant health, and issues of industry regulation. It
is highly unlikely that any international institution has the scope, the resilience, the
political support, or the expertise to provide and support a comprehensive framework for
the regulation of biotechnology (Buckingham and Phillips, 2001).

If in the absence of other institutions - and in the view that most of the disputes
are trade related (market access, SPS measures, technical barriers to trade, etc.) —- the
WTO may be the best suited multilateral body, it might be necessary to re-evaluate its
underlying concepts: end product vs. production method, like products etc. Does a gene
inserted for agronomic reasons change the product, although the consumer experiences
the product as the same? Heumueller and J osling (cited in Sheldon, 2003) suggest that if
the relevant benchmark for deciding a “like good” is the chemical composition of a
product, this would likely undermine domestic regulations on GM foods. To account for
an appended understanding of like goods differentiated (for example) on the basis of
production processes, tariff classifications would have to be substantially appended.

Thus, for the GE foodstuffs case, the EU and US positions suggest potential gains
from trade are not sufficient to justify harmonization, and their respective trading partners

cluster into “clubs” based on similar preferences and actual or perceived gains from trade.

 

6' Starlink is a US corn hybrid approved as a feed. However, in year 2000, Starlink was detected in
manufactured foods.

142

APPENDICES

143

Appendix A: Standards as Strategic Policy Instruments

All numeric examples and solutions were generated with Mathematica’s F indRoot
function using different values for countries’ preference parameters and endowments.
Starting points for numeric evaluations of different endowments and number of iterations
might have differ to achieve the prescribed accuracy of the Newton’s method. Numbers

might be rounded due to computer precision.

A. 1. Social Planner Problem with Identical Endowments

Table A. 1: Social Planner: Numeric Solutions, Stationary Points, Identical Endowments.

 

 

 

 

 

 

 

 

 

Foreign Home Foreign Home Foreign Home Soc Soc
Preference Preference Endwmnt Endwmnt Choice Choice (f) (h)

(F) T”) (Y) TX) (f) (h)

0.5 1.5 2 2 1 1 + +

287.732 287.732 - -

0.5 1 .5 4 4 l 1 + +

287.732 287.732 - -

1 3 2 2 2 2 + +

247.603 247.603 - -

1 3 6 6 2 2 + +

329.742 329.742 - -

l 5 2 2 3 3 + +

226.159 226.159 - -

1 5 10 10 3 3 + +

1 7 10 10 4 4 - -

2.3984 2.3984 + +

5.6016 5.6016 + +

1 9 2 2 5 5 - -

1.7867 1.7867 + +

8.2133 8.2133 + +

1 9 10 10 5 5 - -

1.7867 1.7867 + . +

8.2133 8.2133 + +

Note: F, H, X and Y are exogenous.

More than one stationary point possible, each stationary point is evaluated separately.

144

Analytically:
Home: the second derivative of the social welfare function with respect to Home

choice parameter (h) evaluated at the critical point (h* = (F +H)/2) reduces to:
2
62W _ -64Y[(ln8—2)(F—H) —ln16]

(A.1)
ah2

 

h=h* (“(17-411)2 )3

 

For a critical point to be a local maximum, the second order condition evaluated at the
critical point is negative. For the Equation A.l to satisfy the second order condition to

maximization, it has to be true that

(A.2) (ln8—2)(F—H)2-lnl620
The equation A.2 holds if:
ln16
A.3 F—H 2 ——
( ) ( ) (ln8—2)

The social planer’s decision to split the difference between standards when endowments
across countries are the same is welfare maximizing only in case of the Euclidian
distance between Home and Foreign preference parameters (F and H) being “large”: in
numeric terms the right hand side of the Equation A3 is 5.9. When difference between F
and H is smaller than the critical value, the point in the middle of the interval is welfare
minimizing. Thus, the social planner chooses countries’ preference parameters as
prevailing standards. Same line of reasoning applies for the Foreign country. Graphing
social welfare in a three-dimensional space as a function of Home and Foreign choice

parameters (h and f) yields Figures A.1 and 2.

145

 

Figure A. 1: Social Welfare as a Function of Choice Parameters h, f, Identical

Endowments, Difference between Preference Parameters (H, F) “small”.

Social
ll
If

welfare

(”9

 

Figure A.2: Social Welfare as a Function of Choice Parameters h, j; Identical

Endowments, Difference between Preference Parameters “large”.

146

A.2. Social Planner Problem with Different Endowments

Table A2: Social Planner: Numeric Solutions, Stationary Points, Different Endowments.

Critical Points

 

 

 

 

 

 

 

 

 

 

 

 

 

147

Foreign Home Foreign Home “Foreign “Home Soc Soc
Pref. Pref: Endwmn Endwmn Choice” Choice” (f) (h)
(1:) (H) 0’) (A9 (f) (h)

0.5 1.5 3 2 0.89604 0.84705 + +

0.5 2.5 3 2 1.25785 0.77769 + +

0.5 3.5 3 2 1.55367 0.63070 + +

0.5 4.5 3 2 1.75656 0.57124 + +

0.5 5.5 3 2 1.8065 0.54318 + +

0.5 6.5 3 2 1.6295 0.52819 + +

0.5 7.5 3 2 1.41349 5.8421 + .-

7.5390 0.51945 - +

0.5 8.5 3 2 1.27152 6.2998 + -

8.52 7.68776 - +

0.5 9.5 3 2 1.17562 6.87874 + -

9.52 8.80565 - +

0.51044 +

1 2 3 2 1.39605 1.34705 + +

1 9 3 2 1.77152 1.014 + +

9.02793 6.7999 - -

8.18776 +

1 3 4 2 1.16854 1.13779 + +

1 4 4 2 1.95959 1.0646 + +

1 6 4 2 1.19581 1.02083 + +

l 8 4 2 1.8954 1.00922 + +
8.02767 -

1 10 4 2 1.67208 1.00489 + +

10.0149 7.73795 - -

9.29347 +

1 12 4 2 1.55078 1.00292 + +

12.009 9.03703 - -

1 1.4373 +

1 14 4 2 1.47309 1.00189 + +

14.0059 10.3865 - -

13.5212 +

l 16 4 2 1.418 1.00129 + +

16.0041 11.7566 - -

_ 15.5788 +

1 18 4 2 1.3764 1.00093 + +

18.0029 13.1375 - -

17.6216 +

Table A2: Social planner: numeric solutions, stationary points, different endowments

 

 

 

 

 

 

 

 

 

 

 

 

(cont.).
Foreign Home Foreign Home “Foreign “Home Soc Soc
Preference Preference Endwmnt Endwmnt Choice” Choice” (f) (h)
(F) (H) (Y) (X) (I) (h)
2 3 3 9 2.86036 2.69769 + +
2 4 3 9 3.39247 1.67184 + -
3.37282 +
2 5 3 9 4.96895 1.85228 + -
4.1 1087 +
2 6 3 9 5.98353 1.91862 + -
4.93 549 +
2 8 3 9 7.993 76 1.967 + -
6.97102 +
2 10 3 9 2.93022 1.98325 + -
3.44765 9.2425 -
9.99698 +
2 12 3 9 2.63658 1.99027 + -
4.2324 11.3982 - +
1 1.9983 +
2 20 3 9 2.36186 1.9978 + -
6.53109 19.6411 -
19.9996 +
3 4 2 20 3.96687 3.7572 + +
3 6 2 20 5.999399 2.87724 + -
5.18714 +
3 8 2 20 7.99821 2.95282 + -
6.95781 +
3 10 2 20 9.99924 2.97649 + -
9.14169

Based on Table A2, we induce that when endowments differ across countries, both the
differences in the countries’ endowments (measured by their ratio) and the Euclidian
difference between Foreign and Home preference parameters influence the social
planner’s decision:

1. The difference between countries’ preference parameters and/or the ratio of

countries’ sizes is small, social planner’s selection of choice parameters coincides

148

with the countries’ preference parameters. The social welfare function is convex
on the interval if interest.

2. The difference between countries’ preference parameters is small and/or the ratio
of countries’ sizes approaches a certain critical value, standard of a smaller
country prevail. The social welfare function is neither concave nor convex on the
interval of interest.

3. The difference between countries’ preference parameters and/or the ratio of
countries’ endowments are sufficiently far away from each other, social planner’s
choice of standard for the larger country approaches smaller country’s preference
parameters, while smaller country’s choice standard lies between the preference
parameters, closer to the smaller country’s preference parameter. The social

welfare function is concave on the interval of interest.

Analytical hint of a proof:

As described in the Proposition 2.3, the outcome of the social planner’s
maximization problem depends on the Euclidian distance between the countries’
preference parameters and the ratio of their respective endowments influencing - like in
the case of equal endowments, the curvature of the objective function. Unlike in the equal
case endowments, the different endowments case does not allow for an elegant solution
in solving for a critical value of the Euclidian distance between Home and Foreign
preference parameters and endowments. However, scrutiny of the numerical solutions

together with the first and second order conditions result in useful intuition.

149

The first order conditions coincide with Equations 2.16A and B. The second order

conditions (notationally simplified using Equations 2.2A — 3B) are:

 

2
M=2Xa%{(3(h—H)2-1)ln[ XaH J4-
ahz XaH+YaF

 

3 2 ) XaF
A.4 2r 3 h—F -1 1n +
( ) “A ( ) [XaH+YaF]

“Haifa; (1+(F-h)(h-H))2(F-H)2

 

 

 

 

 

XaH+YaF
23K- 3 _ 2_ Y/3H ]
afZ ’2XflH(3(f H) 1)m[XflH+YflF +
(A's) 2Y3 13” (3U _F)2 -1)ln[XflI:iFYflF]+
4XYB2IB2—(—1+(f-F)(f-H))2(F—H)2
XflH+YflF
2 2
mm 6W=6W=
ahaf afah

The curvature of the social function depends on the signs of the principal minors

of the Hessian matrix. The second order cross-partials (Equation A.6) are zero. Recall by

definition the exponents 01,-, B,- 6 (0,1) (i = F, H). Thus the natural logarithm of weighted

XaH
XaH +YaF

 

consumption shares (ln[ ], etc) is always negative. Both non-zero second

order conditions (Equations A.4 and 5) are similar in structure, and thus we analyze in
detail only one of them. Depending on the position of Home’s choice parameters, the key

150

components with varying signs of the Equation A.4 are (3(h — H )2 — I) and

(3(h — F )2 — I) . While no statements can be made regarding the distance between the

choice parameter and exogenous preference parameters, potential signs of the second
derivative of the social welfare function with respect to Home’s choice parameter are

summarized in Table A3.

Table A3: Sign of the Second Derivative of the Social Welfare Function.

 

 

 

 

 

Sign orig, 3(h-F)2—1<0 3(h-F)2—1=0 3(h-F)2-l>0
3(h—H)2—1<O > 0 > 0 depends
3(h-H)2-1=0 > 0 > 0 depends
3(h— H )2 —1 > 0 depends depends depends

 

 

 

 

 

 

In the fields labeled as “depends” the sign of the second derivative depends on the
distance between the countries’ preference parameters as well as the ratio of their
endowments: in case of small differences the first principal minor of the Hessian matrix
is positive. Similarly we evaluate the second derivative of the social welfare function
with respect to the foreign choice parameter (f). Thus, the signs of the principal minors of
the Hessian matrix cannot be assigned without an ambiguity, and the social welfare
function can be convex, not convex and not concave, or concave.

When the social welfare firnction is convex due to small distance between the

countries’ preference parameters, the objective function cannot be optimized. In this case

151

social planner chooses countries’ respective preference parameters as choice parameters,
as they yield the highest welfare.
When the social welfare function is concave, the local maximum occurs in the

midpoint between the preference parameters.

A.3. Myopic equilibrium, Second order conditions

We employ the second order condition (SOC) to evaluate the SOC whether the stationary

point is a minimum or maximum. The SOC in the Home country is:

2

a VH _ 3 2 [ 2 1 J }
(A.7) —2a 2 h—H 1-6 + 4 h—H — lnB +lnX

ahz H{ ( ) ( H) ( ) a ( H )

Evaluating the SOC in the stationary points described in the text we get

th = -—2[ln|:flHl +l]+m[x]]

_4fl;{(-2+(f-H)2(X—l)+X)

h=H-\/—2-(H-f)2+X+(H—f)2X (X - 02 X

anH
6122

 

(A.8)

 

2
(A.9)y2LI-
ah

 

 

 

(A.10)

02VH _ 4a}, (-2+(f—H)2 (X—1)+X)

ah?-

 

h=H+\/—2-(H-f)2 +X+(H-f)2X (X ‘1)2 X

 

 

First we assign sign to the SOC at the stationary point h = H. For a stationary point to be
a maximum, the SOC evaluated in the stationary point has to be negative. Therefore, it

has to be true that:

152

 

 

(A.ll) ln[ ]+lnX>0
flH +1
which can be rewriten as:
(A 12) In 1 >1n[X]‘1
. 26H +1

Since In is a monotonically increasing function, it has to be true that:

 

 

(A.l3) 1 >21”1 or
flH +1
(A.l4) 5H +1 < X
Recall ,6 H = 2 reaches its maximum value of one at f = H (Figure 2.2).
1+(f—H)

Therefore, if the Home country’s endowment exceeds two (the maximum attainable value
for the left hand side of Equation (A.l3)), h = H is a local maximum. For values of
endowment less than two, h = H would be a local minimum. Thus, we assume

endowments are greater than two.

The sign ofthe A9 and 10 is dependent on(—2+(f—H)2 (X —1)+X) , as the
rest of the expression is positive. It can be rewritten as(—1+(X —1)(( f — H )2 +1».

Since we assumed X to be greater than two, and (f — H )2 is always non-negative, it

follows that (X — 1)(( f - H )2 +1) is a product of two numbers greater than one, and

therefore itself greater than one. Consequently, the sign of the A9 and 10 are positive,

and minimums are observed in the other two stationary points.

153

 

Proof proceeds in the similar fashion for the Foreign country. The SOC in the

Foreign are:

2
(A.lS) a V2——F— =25F{2 (f—F)26F+(—1+3(f—F)2)(ln[l—6F]+lnY)
af

Evaluating the SOC (A. l 5) in the stationary points leads to:

 

 

 

 

 

 

 

2
(A.l6) 3% =-2[1n[1+1 :l+ln[Y]]<0
af f=F 0’F
62,, 4a} (-2+(h—F)2 (Y-1)+Y)
(A. 17) 2F = 2 > 0
51’ f=F—([—2-(F-h)2+Y+(F-h)2 Y (Y ‘1) Y
62V 4a%(—2+(h—F)2 (Y—1)+Y)
(A. 18) 2” = 2 > 0
5f f=F+(/-2—(F-h)2+Y+(F-h)2Y (Y ‘1) Y

 

Following the procedure for Home, we find if the endowment of the Foreign good is

larger than two, maximum occurs in F, while other two stationary points are local

minima.

A.4. Price Behavior

As actual F OCs (rather complex) do not provide any additional intuition, we only
list the FOC conditions evaluated at the border points (F and H), without solving for the
actual stationary point. Taking and evaluating the F OC in the border points for each

country (keeping the other country’s choice parameter constant) we find that:

154

 

 

 

 

(A.19A) if: = 2(H’F)Y >0
dh h=F 1 2 2 2
1+ 1+ H—F X
[ 1+(F—H)’]( ( W
(A.19B) 119—l = (“my <0
d” h=H (l+(F-H)2)X
(A.20A) 5i?— = 2“” (F-HV <0
df f=F (2+(F-h)2)X
(A203) d_p _ 2(F-H)(l+aH)Y >0

 

2

(1+(F—H)2)2[1+(F1-H)2+ap (l+(F—h)2)X

dff=H=

 

Assigning the FOCs of price evaluated in the border points of the interval relies on the
assumption that (H — F) is positive, as H > F, and (F — H) is negative. Therefore, as
Home is trying to optimize (maximize) the price, we find the price function is increasing
at h = F, and decreasing at h = H. A stationary point where the first derivative is zero
must lay between points F and H. Similarly, as Foreign is trying to minimize the price,
the stationary point must lay between F and H: the first derivative evaluated at f = F is

decreasing, while increasing at f = H.

155

A.5.

Nash Equilibrium

Table A.4: Simulations of Nash Equilibrium with Identical Endowments

 

F H X Y h |H - h] f [F — fl

5 1.5 2 2 1 .5 1 .5 lH—hl=lF-fl
5 1.5 3 3 1.30291 .19709 .69709 .19709 |H -h| =|F —fl
5 1.5 4 4 1.35544 .14456 .644564 .144564 |H—h|=|F—f|
5 1.5 5 5 1.37965 .12035 .620349 .120349 |H — h| =|F -f I
5 1.5 6 6 1.394 .106 .605999 .105999 |H-h|=|F- fl
5 1.5 7 7 1.40366 .096343 .596343 .096343 |H-h|=|F— fl
5 1.5 8 8 1.41068 .08932 .58932 .08932 |H - h| =|F - fl
5 1.5 9 9 1.41606 .08394 .583937 .083937 IH—h|=lF-f|
5 1.5 10 10 1.42035 .07965 .57965 .07965 |H - h| =|F — fl
1 2 3 3 1.80291 .19706 1.19706 .19706 [H —h| =|F- fl
1 3 3 3 2.93372 .06628 1.06628 .06628 |H — h| =|F - f|
1 4 3 3 3.97452 .02548 1.02548 .02548 [H - h| =|F - f|
1 5 3 3 4.98798 .01202 1.01202 .01202 IH -h| =|F -fl
1 6 3 3 5.99349 .00651 1.00651 .00651 |H - h| =|F - f|
1 7 3 3 6.99611 .00389 1.00389 .00389 IH — hl =lF - fl
1 8 3 3 7.9975 .0025 1.0025 .0025 |H - h| =|F — fl
1 10 3 3 9.9988 .0012 1.0012 .0012 |H—h|=|F-f|
1 12 3 3 11.9993 .00067 1.00067 .00067 |H -h| =|F -fl
1 14 3 3 13.996 .00041 1.00041 .00041 IH -hl=lF-fl
1 16 3 3 15.9997 .00027 1.00027 .00027 [H — h| =|F — fl
1 20 3 3 19.9999 .00013 1.00013 .00013 lH -h| =|F -fl
1 50 3 3 49.9999 .00001 1.00001 .00001 |H —hl=lF-fl
1 100 3 3 -> 100 90 ->1 ->0 |H-hl=|F— fl
1 5 2 2 4.98084 .01916 1.01916 .01916 |H—h|=|F-fl
1 5 4 4 4.9905 .0095 1.0095 .0095 IH — hl =lF - fl
1 5 10 10 4.9943 .0057 1.0057 .0057 IH - h| =|F - fl
1 5 20 20 4.99562 .00438 1.00438 .00438 |H — h| =|F — fl

156

Table A.4: Simulations of Nash Equilibrium with Identical Endowments (cont.)

 

 

F H X h lH—hl f lF-f I

1 5 50 50 4.99665 .00335 1.00335 .00335 lH —h|=|F-f|

1 5 100 100 4.99716 .00284 1.00284 .00284 lH -h|=|F-fl

1 5 200 200 4.99753 .00247 1.00247 .00247 |H—h|=|F-fl

1 5 500 500 4.99789 .00211 1.00211 .00211 lH—h|=lF-fl

1 5 1000 1000 4.9981 .0019 1.0019 .0019 lH-h|=|F— fl

1 5 106 106 4.99905 .00095 1.00095 .00095 lH—h|=|F- fl

1 5 109 109 4.99937 .00063 1.00063 .00063 |H-h|=lF-f I
Table A5: Simulations of Nash Equilibrium with Different Endowments

F H X Y h IH -hl f IF -fl

.5 1.5 2 3 1.13504 .36496 .730675 .230675 lH—h|>|F- fl

.5 1.5 2 4 1.16112 .33888 .672632 .172632 |H-h|>|F- fl

.5 1.5 2 5 1.17294 .32706 .645095 .145095 |H-h|>|F— fl

.5 1.5 2 6 1.17986 .32014 .628541 .128541 |H—h|>|F- fl

.5 1.5 2 7 1.18449 .31551 .6173 .1173 |H—h|>|F-f|

.5 1.5 2 8 1.18784 .31216 .60907 .10907 |H—h|>|F— fl

.5 1.5 2 10 1.19243 .30757 .597668 .097668 lH-h|>|F- fl

.5 1.5 2 20 1.20183 .29817 .573836 .073836 IH-h|>|F- fl

.5 1.5 4 3 1.34917 .15083 .687949 .187949 IH-h|<|F- fl

.5 1.5 5 3 1.37208 .12792 .683481 .183481 |H—hl<|F— fl

.5 1.5 6 3 1.38611 .11386 .680767 .180767 |H—h|<|F- fl

.5 1.5 7 3 1.39573 .10427 .678916 .178916 IH-h|<|F- fl

.5 1.5 10 3 1.41273 .08727 .65567 .17567 lH-h|<|F— fl

.5 1.5 20 3 1.43364 .06636 .671717 .171717 IH—h|<|F- fl

.5 1.5 50 3 1.44956 .05044 .668741 .168741 |H—h|<|F— fl

1 5 2 3 4.9809 .0191 1.01206 .01206 lH—h|>|F— fl

1 5 4 4.98093 .01907 1.00955 .00955 |H-h|>|F- fl

1 5 6 4.98095 .01905 1.00739 .00739 lH-hl>|F- fl

157

Table A5: Simulations of Nash Equilibrium with Different Endowments (cont.)

 

A.6.

Nash Bargaining Solution

F H X Y h |H - hl f |F - fl

1 5 2 8 4.98096 .01904 1.00636 .00636 IH — h| >|F — fl
1 5 2 10 4.98096 .01904 1.00574 .00574 lH — hl >lF - fl
1 5 2 14 4.98097 .01903 1.00501 .00501 |H - h| >|F - fl
1 5 2 15 4.98097 .01903 1.00488 .00488 lH — hl >|F - fl
1 5 2 20 4.98097 .01903 1.00441 .00441 lH —h| >lF — fl
1 5 2 25 4.98098 .01902 1.0041 .0041 lH — h| >|F - fl
1 5 2 30 4.98098 .01902 1.00388 .00388 lH —h| >|F - fl
1 5 2 40 4.98098 .01902 1.00358 .00358 [H —h| >|F — fl
1 5 2 50 4.98098 .01902 1.00338 .00338 lH -h| >lF - fl
1 5 2 100 4.98099 .01901 1.00287 .00287 |H -hl >|F — fl

Table A6: Simulations of Nash Bargaining Solution with Identical Endowments

 

 

 

F H X Y H W -hl F IF -fl
1 2 2 2 1.26569 .73431 1.73431 .73431 lH-h|=|F-f|
3 3 ->2 0 -)1 0 lH-h|=|F-f|
4 4 -)2 0 ->1 0 |H—hl=|F—fl
6 6 ->2 0 ->1 0 lH-h|=|F-f|
10 10 ->2 0 -)1 0 lH-hl=lF-fl
1 3 2 2 2.87302 .12698 1.12698 .12698 |H—hl=IF-fl
1.9242 1.0758 2.0758 1.0758 |H -h| =IF -f|
1.77136 2.22864 2.22864 1.22864 lH—h|=|F- fl
10 10 ->3 0 ->1 0 lH—hl=lF-fl
20 20 —>3 0 ->1 0 lH-hl=lF-fl
1 4 2 2 ->4 0 ->1 0 lH—hl=lF-fl
3 3 2.60949 1.39051 2.39051 1.39051 lH—hl=lF-fl
4 4 2.50903 1.49097 2.49097 1.49097 lH -h| =|F -f|

158

Table A6: Simulations of Nash Bargaining Solution with Identical Endowments (cont.)

 

 

 

 

F H X Y H lH-hl F lF-fl
1 4 6 6 2.39595 1.60405 2.60405 1.60405 |H —h|=|F-f|
10 10 2.28184 1.71816 2.71816 1.71816 lH—h|=lF-fl
15 15 2.20535 1.79465 2.79465 1.79465 lH -h| =|F -f I
20 20 2.15633 1.84367 2.84367 1.84367 |H—h|=|F-fl
30 30 2.09267 1.90733 2.90733 1.90733 |H—h|=|F— fl
50 50 2.01098 1.98092 2.98092 1.98092 |H-h| =lF- fl
100 100 1.92672 2.07328 3.07328 2.07328 lH-hl =|F— fl
1000 1000 -)4 0 ->1 0 lH—hl=lF-f|
1 5 2 2 95 0 ->1 0 |H-h|=|F-f|
4 4 3.04903 1.95097 2.95097 1.95097 lH-hl =|F- fl
10 10 2.81293 2.18707 3.18707 2.18707 lH -h| =lF-fl
20 20 2.68742 2.31258 3.31258 2.31258 |H—h|=|F— f|
50 50 2.55738 2.44262 3.44262 2.44262 |H -h| =|F— fl
1 5 200 200 2.40509 2.59491 3.59491 2.59491 lH—hl =|F— fl
500 500 2.32238 2.67762 3.67762 2.67762 lH—hl =|F— fl
1000 1000 2.2662 2.7338 3.73379 2.73379 IH-hl =|F— fl
106 106 1.83719 3.16281 4.16281 3.16281 lH-h|=lF-fl
1 6 2 2 ->6 0 ->1 0 lH-h|=lF-fl
3 3.69791 2.30209 3.30209 2.30209 |H-h| =|F— fl
5 3.50333 2.49667 3.49667 2.49667 IH—hl =|F— fl
7 7 3.40542 2.59458 3.59458 2.59458 lH -h| =|F -f|
10 10 3.31801 2.38199 3.38199 2.38199 lH—h|=|F— fl
20 20 3.18023 2.81977 3.81977 2.81977 |H -h| =|F- fl
100 100 2.95226 3.04774 4.04774 3.04774 |H—h| =lF-fl
200 200 2.87711 3.12289 4.12289 3.12289 lH—h| =|F— fl
500 500 2.79106 3.20894 4.20894 3.20894 lH—hl=|F- fl
1000 1000 2.73365 3.26635 4.26635 3.26635 |H—h|=|F-fl
106 106 2.3455 3.6545 4.6545 3.6545 |H-h|=|F— fl

159

Table A.6: Simulations of Nash Bargaining Solution with Identical Endowments (cont.)

F H X Y H lH—hl F lF-fl

 

1 10 2 2 ->10 0 -)1 0 lH-h|=|F-f|
5 5 5.56267 4.43733 5.43733 4.43733 lH—h|=|F- fl

10 10 5.27748 4.72252 5.72252 4.72252 IH - hl =lF - fl

20 20 5.06701 4.93299 5.93299 4.93299 lH-hl =|F- fl

100 100 4.72342 5.27658 6.27658 5.27658 |H —hl =|F — fl

200 200 4.61188 5.38812 6.38812 5.38812 |H-h|=|F— fl

500 500 4.4855 5.5145 6.5145 5.5145 lH—h|=|F- fl

1000 1000 4.40209 5.59791 6.59791 5.59791 |H-h| =lF— f|

106 106 3.86506 6.13494 7.13494 6.13494 |H-h|=|F— fl

 

1 7 2 2 —)7 0 —)1 0 lH—h|=lF-f|
3 3 4.23974 2.76026 3.76026 2.76026 lH-hl=lF- fl

Table A.7: Simulations of Nash Bargaining Solution with Different Endowments

F H X H lH—hl F lF-fl

 

1 2

N

1.93639 .06361 1.66813 .66813 IH —h| <|F -fl
1.8829 .1171 1.65926 .65926 lH—hl <lF- fl
10 1.84053 .15947 1.65222 .65222 |H-hl<|F- fl
20 1.78192 .21808 1.64265 .64265 |H-h|<|F-fl
100 1.70099 .29901 1.63017 .63017 lH-hl<lF— fl
1.65174 .34826 1.6232 .6232 IH-hl <|F— fl

Y
3 1.47175* .52825 1.60417* .60417* lH—h|<|F— fl
5
7

 

5 2.45794 1.54206 2.5024 1.5024 lH -h| >lF -fl
7 2.39562 1.60438 2.51327 1.51327 lH -h| >lF—f I
10 2.34256 1.65744 2.52018 1.52018 |H —h| >lF -— fl
20 2.26313 1.73687 2.52711 1.52711 |H —h| >lF -f I
100 2.14067 1.85933 2.53154 1.53154 lH —h| >lF —f|

«bk-hb-bNNNNNN
M
O
O

160

Table A.7: Simulations of Nash Bargaining Solution with Different Endowments (cont.)

 

 

F H X Y H |H - h| F IF - fl

1 4 4 500 2.05908 1.94092 2.53141 1.53141 |H -h| >lF -fl
4 1000 2.03075 1.96925 2.53093 1.53093 |H -h| >lF — fl
4 106 1.84203 2.15797 2.52391 1.52391 |H —hl >lF —fl

1 10 4 5 5.60371 4.39629 5.35647 4.35647 IH -h| >|F -fl
4 7 5.50724 4.49276 5.39571 4.39571 |H -hl >lF -fl
4 10 5.424 4.576 5.4233 4.4233 lH —h| >|F- fl
4 20 5.29568 4.70432 5.45821 4.45821 |H —h| >lF — fl
4 100 5.09787 4.90213 5.49442 4.49442 |H - hl >lF -fl
4 500 4.96902 5.03098 5.50957 4.50957 |H -h| >lF — fl
4 1000 4.92548 5.07452 5.51348 4.51348 lH -h| >lF— fl
4 106 4.66714 5.33286 5.52691 4.52691 |H -h| >lF- fl

 

* Newton’s method failed to converge to the prescribed accuracy after 2,000,000

iterations.

Second order conditions in Nash equilibrium and Nash bargaining solution are satisfied.

161

Appendix B: Differing Standards, Welfare, and Sub-global Agreements

Table B. 1: Production — Entire Stock of Labor Directed into a Single Quality

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

“Standard” “Beauty”: 63 “Truth”: 6T
Fixed cost 1:3 = 17(53) FT = F(5T)
Marginal cost M3 = M053) MT = M(6T)
Production (Labor B _ B B B T _ T T T
allocation) Ii -F +yi M 11' "F +yi M
Total cost B _ B B B T _ T T T
09:1). wli —wF +wyi M wli —wF +wyiM
B T
piB=——WF +WMB (T: F +wMT
Zero profit yiB yiT
condition B T
(P’=AC’) flini+MB i=i+MT
W Y? w yiT
1 1
piB(l-—)=wMB plT(l-—)=wMT
Profit maximization 0' a
condition piB M B 12,-T M T
(MR.=MC.) "w“ = 1 1 7 = 1 1
(7) (7)
Goods mkt clearing yiB .-. LciB yr‘T = LciT
P,B =PB P,T =PT
Due to symmet
VI 1 1
exp = c3 ciT = cT
B _ T _
Equilibrium output y 1.3 F (a 1) )2? _. F (a 1)
MB MT
B T
n n
Labormarket 215:1, 211:1,
clearing i= 1 i=1 1
Equilibrium B _ L T L
number offrrms n ' F30 n = FTa

 

162

 

B. 1. Tradeoff between Number of Varieties and Price: “Quantity and Quality”

To allow for general discussion of a tradeoff between the price and number of
varieties available under each attribute (if the entire stock of labor is directed into one
standard only), assume that the marginal labor component of Truth is more than the
marginal labor component of Beauty (M > MB), and the fixed labor component of Truth
is less than the fixed labor component of Beauty (F T < F8). The quality associated with
the lower marginal cost (Beauty) will have the lower price, and lower fixed costs (Truth)
lead to more varieties. Combining both fixed and marginal costs, per firm output of a
variety made in Beauty specification exceeds per firm output of Truth specification.

Accurate selection of the standard depends on the relationship between the
individual country’s discount parameter and a function of cost differential — defined by
underlying technology. Consumers compare indirect utilities62 obtainable from bundles
consisting of different number of varieties depending on quality. Possible bundles include
Truth Only, a mixture of Truth and Beauty (but consuming only one quality per variety
due to linear sub-utility function), or a bundle containing Beauty Only“. We show that a
“mixed and matched” bundle is always dominated by one of the “pure” bundles. Owing
the pricing symmetry — all varieties with same attribute are priced equally — the indirect
utility function from consuming the Truth Only (TO) bundle is:

1—61
(13.1) V10 =LlnTl7

pT

Similarly, from Mix and Match (MM):

 

‘2 Social welfare consists of consumers’ indirect utility, as profits equal zero.

63 We assume a consumer possesses a perfect knowledge about all potentially available varieties, and thus
are able to make indirect utility comparisons between alternatives encompassing different number of
varieties. An opposite would be the case when a consumer fist chooses quality for the varieties he is sure
off, and then considers additional varieties previously not available to him.

163

i B n
W1)" [6:]

 

 

And finally the indirect utility from consuming Beauty Only (BO) bundle is:
1_-Q
(13.3) V30 =iln3) 6

A
pB

Notice we dropped the superscript on A to derive autarky standard selection
criteria valid for both Home and Foreign countries. In the above superscript denotes type
of consumption bundle, subscript indicates autarky, pT ande stand for price of each
Truth and Beauty made variety, respectively. nT and n3 represent number of varieties

available when the entire stock of labor is directed to a single standard based on Equation

3.12. h T and h B label number of varieties available when the stock of labor is divided to
produce some varieties in Truth and some in Beauty. Due of the fixed amount of labor
available in the economy, the number of varieties consumed in the Truth Only bundle
exceeds the number of varieties under the Mix and Match (M&M), which in turn is larger

than number of Beauty Only varieties:

(13.4) nT>ZT+ZB>nB

L is the stock of labor available, and label L8 as labor used to produce Beauty-made
varieties. ; T} B are defined as:

-T L-LB
n =

(B.5A)
F To

 

164

_ B
(3.58) n B = -LT
F 0'

First we derive a condition on ,1 for Truth Only to be preferred over M&M, or:

T0 MM
(B.6) VA > VA

After solving for a condition for A in terms of consumer’s perceived parameters we get:

 

1:9

pB nT_;T 0
(8.7) 2. < T _ B , which corresponds to:

P n

1—6’
B B T

(3.8) A<L[£_]

MT FT

M&M dominates Beauty Only if the indirect utility from consuming an assorted bundle

exceeds the indirect utility from Beauty Only:

 

 

(13.9) V 34M > V fa, which holds if:
1.:6:
B ET 9
(8.10) l < p _ or, in cost terms:
P nB-n
1-6
B B 7
(8.1 1) ,1 < M T F—T—
M F

Notice the condition for Truth Only to be preferred to M&M (3.8) is identical with the

condition for M&M to be preferred over Beauty Only (B.11). By the virtue of transitivity:

165

1—0

(13.12); <

 

MB FB
MT FT

7
] :>(TO>MM)A(MM>BO)=>TO>BO:>TRUTH
Otherwise Beauty prevails. Different qualities are substitutes in consumption: rather than
putting resources into two different makes of the same variety, utility can be increased by
consuming additional varieties. The M&M bundle is always dominated by one of the
“pure strategies” with a uniform standard across all varieties consumed. The autarky

equilibrium is identical with Krugrnan (1980). Define “A” such that:

1-0
B B ‘—
(B.13) AEM [F )9

M T F T
Substitute to Equation B. 12 to obtain the following decision rules:
(8.14) 2. < A for Truth to prevail as autarky standard

(B. 1 5) A > A for Beauty to prevail as autarky standard

In section 3.2 (model with identical fixed and marginal costs across varieties) the

critical value was A = 1. Since by definition (from Section 3.1) ,1 H < l and A F > 1,
Home always chose Truth, and Foreign always chose Beauty.

The trade-off between number of varieties and their price adjusts the cutoff point
(Equation B. 13). The critical value for Truth or Beauty to prevail is a function of cost
differentials of both marginal and fixed costs. Due to assumptions placed on cost shares,

l-B
(FE/FT)? >1 and MB/MT < 1, it is ambiguous whether “A” is greater or
smaller than one. Thus, we recognize two separate cases: A < 1 and A > 1 . Table 3.1

shows autarky prevailing standards in each country (assuming in borderline cases when a

166

country is indifferent between standards, it chooses standard prevailing in the other

country to facilitate potential open trade).

Table 8.2: Autarky Prevailing Standards in Each Country

 

 

 

AUTARKY A >1 A <1
Home AH e (O , A) Truth prevails
AH <1 Truth always prevails in autarky H
A e (A ,1) Beauty prevails
Forleign AF 5 (1’ A) Truth prevails Beauty always prevails in
(2' >1) 1F 6 (A,oo) Beauty prevails autarky

 

 

 

 

 

Open trade recommences depending on what standards prevail in autarky in each

country (summarized in Table B.l). When identical autarky standard prevails in each
country (A > 1 anle e(1,A); orA <1anth e<A,1)), we obtain the usual Krugrnan

(1979, 1980) open trade result where each country benefits from a larger number of
varieties at own (autarky) preferred standard.

A novel case to analyze is when different countries choose different standards

(AF > A >1; or AH < A <1): Home chooses Truth, while Foreign chooses Beauty as
their autarky prevailing standard. After adjusting for critical value of A (earlier A = 1),
Figures 3.3 — 11 apply: non-standardized trade always dominates autarky, while
standardized trade may be welfare worsening. Table B.2 summarizes numbers of
varieties produced under autarky and harmonization from the perspective of the
harmonizing country based on Equation 3.12. L is the amount of labor available in
autarky, and y represents the increase in L. At the moment we assume countries are of the

same size, and observe the same increase in number of varieties available.

167

Table 8.3: Numbers of Varieties in Autarky Unilateral Harmonization

 

 

 

 

r Autarky number of varieties with all labor ,, T =__L_
n available allocated to Truth F To-
Autarky
B Autarky number of varieties with all labor n B =_l;_
n available allocated to Beauty F Bo-
=T Integrated number of varieties with all 2' = 7L
Complete n labor available allocated to Truth F T a
Unilateral
Harmonization = B Integrated number of varieties with all : = 7L
n labor available allocated to Beauty F B 0'

 

 

 

 

 

 

With unilateral harmonization the number of varieties consumed in open trade
increases (in absolute value) y times compared to autarky. For Home to completely
unilaterally adjust to Beauty, the indirect utility from consuming integrated number of
Beauty-made varieties must exceed the utility gained from consuming autarky number of
T ruth-made varieties. This happens if

1:42

(3.16) 2” >0] 9 A

7’
Similarly, for the Foreign country to unilaterally adjust to Truth, it must be true that

1—_6
(B. 1 7) 2. F < y 9 A
where y represents the increase in labor available relative to autarky. When y = I , the
selection rules correspond to the autarky selection rule derived in B.12. The cutoff point
depends on the relationship between the discount factor (,1) and the ratio of fixed and

marginal costs adjusted for the elasticity of substitution (or, in consumers’ terms, a

tradeoff between prices and numbers of varieties) and scaled by y.

168

B.2. Existence of Equilibrium in Open Trade with No Harmonization Required

Cases when the consumption bundle consists of varieties made of one
specification only (autarky or harmonization) coincide with Krugrnan (1980), treating the
“discount parameter” (A) as a monotonic transformation. Therefore, we only address the
existence of equilibrium in open trade with no harmonization required.

We define a heterogeneous good by its variety and attribute, (i, Q), where j
indexes the variety, and Q indexes the attribute. Given two possible standards, Truth (T)
and Beauty (B), L consumers in each country (index domestic consumers 1 to L, and
foreign consumers L+1 to 2L), n” firms in the Home country and nF firms in the Foreign

country each producing a differentiated variety, open trade equilibrium with no

harmonization required is defined to be a price vector p* e Rf. (e — the dimension —

defined as a sum of all T ruth- and Beauty-made varieties in the open economy,

homogenous good, and wage), a consumption allocation (dug *) consisting of d UQ *
units of the heterogeneous good (i, Q) allocated to consumer i,
. . H F . .
V1 6 (1,...,L,...,2L),} e 1,...,n ,...,n ,Q 6 (B,T) and production allocations
Q ... . H F . . .
y j V} e 1,...,n ,...,n ,Q 6 (B,T), such that the followrng conditions are

satisfied:

1. Firms’ zero profit and profit maximization conditions: for each firm j producing a

differentiated product, yJQ,Q = T v Q = B solves:

(8.18) p]. * = ACJ-Q * Vj e (1,...,nH,...,nF) A (Q = T v Q = B) (Zero profit)

169

(3.19) MCJQ'“ = MRJQ * Vj e (1,...,nH,...,nF) A (Q = T v Q = B) (Profit max.)
2. Utility maximization: for each consumer i (i e (1,...,L,...,2L)) in each country -

O (0 = H v 0 = F), (xflcJQ *j‘v’j e (1,...,nH,...,nF),Q e (B,T) solves:

1
nH g "F 0 B 9 6
Max UOT— (do!) + Z (2. dOk)
H 1:1 B k=nH+1
(13.20) {dr}" {d3 ” x
0} j=1, 0" j=nH+1
"H F
T T B B _
J: k=n +1

(For transparency we index differentiated goods produced at Home by the
notation j, and differentiated goods produced in the Foreign country by the notation k.)
3. Markets clear:

0

n
(3.21) X [)9 = L (Labor market)

i=1

"H "F

T T _ B B

(B22) 2 p j T j - X p k Tk (Balanced trade)

1:] k=n +1
(3.23) yJT- = Ldfij + T f y f = Ldfk + T]? (Differentiated goods)
(13.24) T f = Lat)?!- TkB = Ldflk (Imports’ distribution)

170

We now show all equilibrium conditions (B. 1 8 — 24) are satisfied. Normalize

wage to one. As number of firms in a perfectly competitive sector is indeterminate,

assume one firm. Derive demand functions for the differentiated good (0 = H v 0 = F):

 

 

(PT-T”
T _ 1
(B25) d0]. - 1_0 a
H F B
n 1—0' n p
T k
(”1) + Z 0
_ H A
1’1 k=n +1
(103)”
B _ K
(B.26) d0.— H F a
2W1 p!) + 2 (pk)
1:1 k=nH+l

The analyses generally follow Krugrnan (1980) treatment of transport cost with two
exceptions: (1) countries in this model are of equal size, and (2) 1.0 is country variant.
Since the real prices to consumers of different qualities in the Home and Foreign
countries will not be the same due to different weights assigned consumption of different

qualities, we define a set of “real” prices paid by consumers:

__ B
P
(3.27) p 11:3 = —;‘I— for Home consumers purchasing Foreign (Beauty) varieties, and

11

(B28) p3: = A Fp? for Foreign consumers purchasing Home (Truth made) varieties.

Due to real price differences between the countries, consumption of each
imported variety (of a different quality) will differ from consumption of the domestic

made varieties. Dividing Equation 3.25 by B.26 we found each Home consumer will

171

T U B O' H l-O’ . . .
consume ( p j ) ( p k ) (A. ) umts of a representative imported good for

each unit of a representative domestic good he consumes. To explore impact on the
producers, consider “adjusted” demand for the exported good: that is, demand perceived

by consumers:

_ B
d
(B29) d 131k = J? for Foreign producers exporting to the Home country and
( ”)
ll

— 6
(B30) 61;]- = (,1 F ) dig]. for Home producers exporting to the Foreign countr'y

Taking into account adjusted demand, we construct a ratio of total demand by domestic

consumers for each foreign product to demand for each domestic product:

N- 0 L2
(8.31) a): -_—L- (AH)0’ forthe Home countryand
pf
B a _1__2
(B32) (0*: ilk—f- (217)0' forthe Foreign country
1’1

Consumer, having access to varieties not available in autarky, exhaust his entire

differentiated goods budget such that:
(B.33) (n HpT + am Fp B)dT =1 where air is the consumption ofthe
representative domestic good. Similarly in the Foreign country:

(334) [w*anT+anB]dB=l

172

The elasticity of export demand facing any given firm is identical with the elasticity of
domestic demand (a). The autarky conclusions regarding the production are valid:
production in the differentiated sector of each country (Home producing Truth-made
only, Foreign producing Beauty-made only) is derived in Appendix B]. Due to

symmetry all goods of the same quality will be produced in the same quantity and at the
same price: shorthand the notation: p; = pT,p£ = pBandyZ: = yT,yf = yB. Due

to constant elasticity from the profit maximization condition itself we obtain profit

maximizing price:

(8.36) p9 =—“——MQ
a—l

Combine the profit maximizing and zero profit conditions to obtain equilibrium output:

(337) yQ = flifll
MQ

From the labor market clearing condition (321) derive equilibrium number of firms

L
B.38 n = —
( ) Fa

The equilibrium number of firms and nominal prices in each country remain the same.
Nevertheless, the model as specified with country-variant discount factors (As) does not
exhibit identical real wages even in autarky“. Once the trade opens, there exists a relative
real wage ratio between Home and Foreign countries such that the trade is balanced.

From Equation 3.31 and 32 an Open economy real wage is defined as:

 

a
6‘ Autarky real wage in the home country - assuming Truth prevails — is (l/ p T) , while real wage in the

a
Foreign — Beauty loving country — is(h F / p B ) , where 1F > 1. Therefore, real wage in the Foreign

country is always less than the real wage in the Home country.

173

 

(3.39) r = 1

1‘0 a—l
(pita £31
10

K )

 

 

where 0 is H (Home) or F (Foreign). Define a relative real wage ratio such that:

(B40) 1;! = —

Combine B33 and 34 to find the balanced trade condition: for the trade to be balanced it

has to be true that:

(3.41) nwadT =na)*pTdB,
which, multiplied by the number of workers in each economy leads to the Home’s

balance of payments condition (measured in real wage units of the Foreign country):

 

*
(3.42) 3=ny[ ‘0 — w ]
w*+1 w+1

Using the Krugman (1980) argument, since a) and to" are both functions of w, 3.42 can

be used to determine open trade real relative wage ratio. At the equilibrium open trade
real relative wage ratio (if) the expression in the brackets in B42 is zero, and trade is
balanced. Since an is increasing in w, and to" is increasing in w, B(w) will be negative
(positive) if and only if w is greater (less) than J , which shows that J is the unique open

trade relative real wage.

174

BIBLIOGRAPHY

. Andiamananjara, S. 1999. On the Size and Number of Regional Integration
Arrangements: A Political Economy Model. College Park, MD, University of
Maryland.

. Appleton, A.E. 1999. “Shrimp/Turtle: Untangling the Nets”, Journal of International
Economic Law, Sept: 477-496.

. Baldwin, R. 1993. A Domino Theory of Regionalism. Cambridge, MA, National
Bureau of Economic Research.

. Bemauer, T. 2003. Genes, Trade, and Regulation. Princeton and Oxford, Princeton
University Press.

. Bhagwati, J ., A. Panagariya, and T.N. Srinivasan. 1998. Lectures on International
Trade. Cambridge, MA, The MIT Press.

. Bhalla, Raj and Kevin Kennedy. 1998. World Trade Law: The GA TT- WTO System,
Regional Arrangements, and US. Law. Charlottesville, Virginia: Lexix Law
Publishing.

. Bond, E.W. and C. Syropoulos. 1996. “The Size of Trading Blocs: Market Power and
World Welfare Effects”, Journal of International Economics May: 411 - 37.

. Brander, IA. 1995. Strategic Trade Policy. In Handbook of International Economics,
vol. III (eds. Grossman and Rogoff): 1395 - 1455. Amsterdam; New York and
Oxford: Elsevier, North-Holland.

. Brauer, J. and A. Roux. 2000. “Peace as an International Public Good: An
Application to Southern Africa”, Defense and Peace Economics 11(6): 643-59.

10. Buchanan, J. M. 1965. “An Economic Theory of Clubs”, Economica, New Series

32(125): 1-14.

11. Buckingham, DE. and P.W.B. Phillips. 2001. “Hot Potato, Hot Potato: Regulating

Products of Biotechnology by the International Community”, Journal of World Trade
35(1): 1 — 31.

12. Carraro, C. and D. Siniscalso. 1993. “Strategies for the International Protection of the

Environment”, Journal of Public Economics 52(3 (October)): 309-28.

13. Carter, CA and GP. Gruere. 2003. “Mandatory Labeling of Genetically Modified

Foods: Does It Really Provide Consumer Choice?” AgBioForum v6, n1-2: 67 - 70.

175

14. Carter, C.A. and GP. Gruere. 2003. “International Approaches to the Labeling of
Genetically Modified Foods”, Choices Second Quarter: 1 — 4.

15. Casella, A. and J. S. Feinstein. 1990. Public Goods in Trade: On the Formation of
Markets and Political Jurisdictions. Cambridge, MA, NBER.

16. Casella, A. 1992. “On Markets and Clubs: Economic and Political Integration of
Regions with Unequal Productivity”, The American Economic Review 82(2): 115-
121.

17. Casella, A. 2001. “Product Standards and International Trade: Harmonization through
Private Coalitions?” Kyklos v54, n2-3: 243 - 64.

18. Caswell, Julie A. 2000. “As Evaluation of Risk Analysis as Applied to Agricultural
Biotechnology (With a Case Study of GMO Labeling)”, Agribusiness 16(1):115-123.

19. Codex Alimentarius. 2004. Various materials available online at
http://www.codexe_11imentarius.net

20. Comes, R. and T. Sandler. 1996. The Theory of Externalities, Public Goods, and Club
Goods. Cambridge, UK, Cambridge University Press.

21. Darby, MR. and E. Karni. 1973. “Free Competition and the Optimal Amount of
Fraud”, Journal of Law and Economics v16, n1: 67 — 88.

22. Dasgupta, A. 2000. “Labour Standards and WTO: A New Form of Protectionism”,
South Asia Economic Journal v1, n2: 113 — 29.

23. David, P. 1985. “Clio and the Economics of QWERTY”, American Economic Review
75, n2: 332 — 7.

24. Dall’Olio, A. and R. Vohra. 2001. “Efficiency in an Economy with Fixed Costs”,
Journal of Public Economic Theory 3: 185 — 201.

25. Dixit, Avinash and Joseph Stiglitz. 1977. “Monopolistic Competition and Optimum
Product Diversity”, American Economic Review 67(3): 297-308.

26. Donovan, J .A., J .A. Caswell, and E. Salay. 2001. “The Effect of Stricter Foreign
Regulations on Food Safety Levels in Developing Countries: A Study of Brazil”,
Review of Agricultural Economics v23, n1: 163 — 75.

27. European Commission (2004). Various online references.

28. Ederington, J. 2001. “International Coordination of Trade and Domestic Policies”,
American Economic Review v91, n5: 1580 - 93.

176

 

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

Ederington, J. 2002. “Trade and Domestic Policy Linkage in International
Agreements”, International Economic Review v43, n4: 1347 - 67.

Eggers, Barbara and Ruth Mackenzie. 2000. “The Cartagena Protocol on Biosafety”,
Journal of International Economic Law Vol 3(3): 525-43.

Emerson, M. et al. 1988. The Economics of I 992: The E. C. Commission’s Assessment
of the Economic Eflects of Completing the Internal Market. Oxford; New York;
Toronto and Melbourne: Oxford University Press.

Emerson, RM and AC. Nessman. 1995. “Integrating Environmental Protection and
North American Free Trade”, Review of Marketing and Agricultural Economics v63,
n2: 243 — 55.

Esty, DC. 1994.” Making Trade and Environmental Policies Work Together: Lessons
from NAFTA”, Aussenwirtschaft v49, n1: 59 - 79.

European Union in the US (2003). GMOs: Commission Publishes Recommendations
to Ensure Co-existence of GM and non-GM Crops. News Release No. 46/03, July
23, 2003.

Foreign Agricultural Service, US Department of Agriculture. Foreign Agricultural
Trade of the United States. Available on-line at
http://www.fas.usdagoWustrade/USTExFatus.as ? I=

Fox, J. L. 2003. “Agbiotech climbs Afiica’s agenda”, Nature Biotechnology
21(6):589.

F ratianni, M. and J. Pattison. 2001. “International Organizations in a World of
Regional Trade Agreements: Lessons from Club Theory”, World Economy 24(3):
333-358.

Gabszevvicz, J ., A. Shaked, J. Sutton, and J .F. Thisse. 1981. “International Trade in
Differentiated Products”, International Economic Review 22: 527-535.

Gaisford, J. D. and W. A. Kerr. 2001. Economic Analysis for International Trade
Negotiations. Cheltenham, UK, Edward Elgar.

Gaisford, J .D. 2002. “Agricultural Biotechnology and the FTAA: Issues and
Opportunities”, Estey Centre Journal of International Law and Trade Policy v3, n2:
328 - 45.

Gandal, N. and O. Shy. 2001. “Standardization Policy and International Trade”,
Journal of International Economics v53, n2: 363 — 83.

Gao, Lanrong, 2003. China becomes one of the top transgenic planting countries.

177

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

People’s Daily, Dec 07 2003, available on-line at
http://english.people.com.cn/200312/07/eng20031207 129867.3htm1

Garbo, L. 2002. “International Convergence in Regulation: The Case of Measurable
Standards”, International Advances in Economic Research v8, n1: 10 — l9.

Griffin, K. 2000. “Delinking Trade, Environmental Protection, and Labor Standards”,
Environment and Development Economics v5, n4: 504 - 06.

Grossman, G. M. and E. Helpman 1995. “The Politics of Free Trade Agreements”,
The American Economic Review 85(4): 667-690.

Helpman, E. and P. Krugman. 1985. Market Structure and Foreign Trade.
Cambridge, MA: MIT Press.

Henson, S. and R. Loader. 1999. “Impact of Sanitary and Phytosanitary Standards on
Developing Countries and the Role of the SPS Agreement”, Agribusiness v15, n3:
355 — 69.

Hoekman, B., M. Schiff, et a1. 1998. Regionalism and Development: Main Messages
from Recent World Bank Research. Washington, DC, World Bank.

Hufbauer, G., B. Kotschwar, and J. Wilson. 2002. “Trade and Standards: A Look at
Central America”, World Economy v25, n7: 3 — 16.

International Service for the Application of Agricultural Biotechnology (ISAAA).
CropBiotech Update, Feb 27, 2004. Available on-line at http://www.isaaaorg

Isaac, Grant E. and William A. Kerr. 2003. “Genetically Modified Organisms at the
World Trade Organization: A Harvest of Trouble”, Journal of Trade Law 37(6):
1083-1095.

James, C. 2003. Preview: Global Status of Commercialized Transgenetic Crops:
2003. ISAAA Briefs No. 30. ISAAA: Ithaca, NY.

Keohane, R. O. and J. Joseph S. Nye. 2001. “Between Centralization and
Fragmentation: The Club Model of Multilateral Cooperation and Problems of
Democratic Legitimacy”, KSG Working Paper No. 01-004.

Kerr, W. A. 1999. “International Trade in Transgenetic Food Products: A New Focus
for Agricultural Trade Disputes”, World Economy.

Kerr, W. A. 2000. “Is it Time to Re-think WTO? A Return to the Basics”, The Estey
Centre Journal of International Law and Trade Policy 1(2): 99-107.

Kerr, W. A. and A. L. Hobbs. 2001. “Taming the Dragon: The WTO afier the

178

 

57.

58.

S9.

60.

61.

62.

63.

64.

65.

66.

67.

68.

69.

Accession of China”, The Estey Centre Journal of International Law and Trade
Policy 2(1): 1-9.

Kerr, W. A. 2002. “A Club No More - The WTO afier Doha”, The Estey Journal of
International Law and Trade Policy 3(3): 1-9.

Kerr, W. A. 2003. “Science-based Rules of Trade — A Mantra for Some, An
Anathema for Others”, The Estey Journal of International Law and Trade Policy
4(2): 86-97.

Kindleberger, C. P. 1981. “Dominance and Leadership in the International Economy:
Exploitation, Public Goods, and Free Rides”, International Studies Quarterly 25(2):
242-254.

Klevorick, AK. 1996. Reflections on the Race to the Bottom. In Fair trade and
harmonization: Prerequisites for fiee trade? Volume 1. Economic Analysis: 459 — 67.
Cambridge and London: MIT Press.

Krugrnan, P. 1979. “Increasing Returns, Monopolistic Competition, and International
Trade”, Journal of International Economics 9(4): 469-79.

Krugrnan, P. 1980. “Scale Economies, Product Differentiation, and the Pattern of
Trade”, American Economic Review 70(5): 950-959.

Krugrnan, P. and M. Obstfeld 2002. International Economics: Theory and Policy (6“
Edition). Pearson Addison Wesley.

Krugrnan, P. and A. Smith. 1994. Empirical Studies of Strategic Trade Policy. A
NBER Project Report. Chicago and London: University of Chicago Press.

Lai, E.L.-C and L.D. Qiu. 2003. “Intellectual Property Rights Protection and
Endogenous Economic Growth”, Journal of Economic Dynamics and Control v27,
n5: 853 — 73.

Laird, S. 2001. “Dolphins, Turtles, Mad Cows and Butterflies - A Look at the
Multilateral Trading System in the let Century”, World Economy.

Lawrence, R. Z. 1996. Regionalism, Multilateralism, and Deeper Integration.
Washington, DC, The Brookings Institution.

Levinson, A. 1996. Environmental Regulations and Industry Location: International
and Domestic Evidence. In Fair trade and harmonization: Prerequisites for free

trade? Volume]. Economic analysis: 429 — 57. Cambridge and London: MIT Press.

Levy, P. I. 1997. “A Political-Economic Analysis of Free Trade Agreements”, The
American Economic Review 87(4): 506-519.

179

70.

71.

72.

73.

74.

75.

Lindenmayer, I. 1999. “Harmonization of Food Regulations and Food Quality /
Safety Measures Based on Codex Standards, Guidelines and Recommendations.”
Conference on International Food Trade Beyond 2000: Science-Based Decisions,
Harmonization, Equivalence and Mutual Recognition. Melbourne, Australia, 11 — 15
October 1999. Available online at http://www.fao.org/docrep/meeting/X2809E.htm .
Last viewed on July 8, 2004.

Lutz, S. 1993. Vertical Product Diflferentiation, Minimum Quality Standards, and
International Trade Policy. Dissertation. Purdue University.

Lynch, D. 2000. “Closing the Deception Gap: Accession to the European Union and
Environmental Standards in East Central Europe”, Journal of Environment and
Development v9, n4: 426 - 37.

Maddison, A. 1995. Monitoring the World Economy, 1820-1992. Paris, OECD.
Maddison, A. 2001. The World Economy. Paris, OECD.

Madkour, M.A., A.S. El Nawawy, and PL. Traynor. 2000. Analysis of a National
Biosafety System: Regulatory Policies and Procedures in Egypt. ISNAR Country
Report No. 62. The Hague: International Service for National Agricultural Research.
Available at
http://www.essentialbiosafety.info/cstudies.php?book=ISNAR&ev=COUNTRYRPT6
_2_, views on June 23, 2004.

76. Maredia, M., F. Erbisch, A. Naseem, A. Hightower, J. Oehmke, D. Weatherspoon,

77.

78.

79.

80.

81.

and C. Wolf. 1999. “Public agricultural research and the protection of intellectual
property: issues and options”, AgBioForum, 2(3&4), 247-252. Available online at
http://wwwagbioforumorg.

Mas-Colell, A., MD. Whinston, and J .R. Green. 1995. Microeconomic Theory.
Oxford, Oxford University Press.

Mattoo, A. 2001. “Discriminatory Consequences of Non-Discriminatory Standards”,
International Journal of Economic Integration v19, n1-2 (January): 55 — 77.

McNelis, N. 2000. “EU Communication on the Precautionary Principle”, Journal of
International Economic Law Sep: 545 - 551.

McNellis, N. 2001. “The Role of the Judge in the EU and WTO”, Journal of
International Economic Law March: 189 - 208.

Millstone, E., E. Brunner, and S. Mayer. 1999. “Beyond ‘Substantial Equivalence’”,
Nature, October 7, 1999.

180

82. Nielsen, C.P, S. Robinson, and K. Thierfelder. 2002. Trade in Genetically Modified
Food: A Survey of Empirical Studies. Washington, DC. IFPRI. TMD Discussion
Paper no. 106.

83. OECD 1996. Regionalism and its Place in the Multilateral Trading System. Paris,
OECD.

84. OECD 2003. Agra-Food Products and Technical Barriers to Trade: A Survey of
Issues and Concerns Raised in the WT 0 ’s T BT Committee. Paris, OECD.

85. Oehmke, J. F .; M. Maredia and D Weatherspoon. 2001. “The Effects of
Biotechnology Policy on Trade and Growth”, International Law and Trade Policy
Vol 2 (No. 2):283-306. Available on-line at hmazflwwwestg/ioumalcom

86. Olson, M. J. and R. Zeckhauser. 1966. “An Economic Theory of Alliances”, The
Review of Economics and Statistics 48(3): 266-279.

87. Olson, M. J. 1968. The Logic ofCollective Action. New York, Schocken Books.

88. Pauwelyn, J. 1999. “The WTO Agreement on Sanitary and Phytosanitary (SPS)
Measures as Applied in the First Three SPS Disputes”, Journal of International
Economic Law Dec: 641 - 664.

89. Perdikis, N. 2000. “A Conflict of Legitimate Concerns or Pandering to Vested
Interests? Conflicting Attitudes towards the Regulation of Genetically Modified
Goods—The EU and the US”, International Law and Trade Policy, Vol 1. Available
on-line at h@://www.esteviournal.com

90. Perdikis, N., W. A. Kerr, et al. 2001. “Reforming the WTO to Defuse Potential Trade
Conflicts in Genetically Modified Goods”, World Economy.

91. Phillips, P.W.B. and WA. Kerr. 2000. “Alternative Paradigms: The WTO versus the
Biosafety Protocol for Trade in Genetically Modified Organisms”, Journal of World
Trade Vol. 34(4): 63-75.

92. Pimbert, M. 1997. “Issues Emerging in Implementing the Convention on Biological
Diversity”, Journal of International Development 9(3 (May - June 1997)): 415-25.

93. Qaim, M. and A.de Janvry. 2003. “Genetically Modified Crops, Corporate Pricing
Strategies, and Farmers' Adoption: The Case of Bt Cotton in Argentina”, American
Journal of Agricultural Economics. Vol. 85 (4): 814-28.

94. Rege, V. 2002. “Theory and Practice of Harmonization of Rules on Regional and
Multilateral Basis: Its relevance for World Trade Organization Work on Trade
Facilitation”, Journal of World Trade 36(4): 699 — 720.

181

 

 

95. Republic of Namibia, Ministry of Trade and Information. Trade Preference Schemes.
Available on-line at
http://www.mti.gov.na/tradeopps_text/trade_preference_schemestnamibia.htm

96. Richardson, M. 1994. “Why a Free Trade Area? The Tariff Also Rises”, Economics
and Politics v6, n1 (March): 79 - 96.

97. Richardson, J .B. 2000. “EU agricultural policies and implications for
agrobiotechnology”, AgBioForum 3(2&3), 77-83. Available online at
http://www.agbioforum.oLg

98. Runge, CF. and LA. Jackson. 2000. “Labeling, Trade and Genetically Modified
Organisms. A Proposed Solution”, Journal of World Trade 34(1): 111 — 122.

99. Runge, C. F., G.-L. Bagnara, et a1. 2001. “Differing US. and European Perspectives
on GMOs: Political, Economic and Cultural Issues”, The Estey Centre Journal of
International Law and Trade Policy 2(2): 221-234.

100. Sampson, G. P. 2000. “The World Trade Organization after Seattle”, World
Economy.

101. Schiff, M. and L. A. Winters. 1997. Regional Integration, Security and Welfare.
Washington, DC, World Bank.

102. Shapiro, C. and H. Varian. 1999. Information Rules. Harvard Business School
Press: Boston, MA.

103. Sheldon, I. 2003. Europe’s Regulation of Agricultural Biotechnology: Precaution

or Protection. Paper presented at conference on “The Role of GMOs in Trade
Disputes”, San Antonio, TX, October 30-31, 2003.

104. Singh, P. 2002. “NAFTA and Labor: A Canadian Perspective”, Journal of Labor
Research 23(3): 433 -— 46.

105. Sykes, A. 1999. “The (Limited) Role of Regulatory Harmonization in
International Goods and Services Markets”, Journal of International Economic Law
Jan: 49 — 70.

106. Tirole, J. 1988. The Theory of Industrial Organization. The MIT Press.

107. Tothova, M. and J .F. Oehmke. 2004. “Genetically Modified Food Standards as
Trade Barriers: Harmonization, Compromise, and Sub-Global Agreements”, Journal
of Agricultural and Food Industrial Organization: Vol. 2: No.2, Article 5.
http://www.bepress.com/iafio/volZ/iss2/art5

108. Venables, A. J. 2000. International Trade, Regional Economic Integration.

182

 

International Encyclopedia of Social and Behavioral Sciences. A. J. Venables.

109. Wilson, John S. 1997. International Trade: Standards, Technical Regulation, and
Global Reform. Available online at
http://siteresources.worldbank.org/lNTRANETTRADE/Resources/Marrakech_Lectur
gpclf . Last viewed on July 7, 2004.

110. Wilson, John S. 2001. Bridging the Standards Divide: Recommendations for
Reform from a Development Perspective. World Bank. Available online at
http://wwwl .worldbank.orglwbiep/trade/Standards/files/BPforWDR.pdf

111. Wilson J .D. 1996. Capital Mobility and Environmental Standards: Is There a
Theoretical Basis for a Race to the Bottom? In Fair trade and harmonization:
Prerequisites for free trade? Volume I . Economic Analysis: 393 - 427. Cambridge
and London: MIT Press.

112. WTO. 1995. Regionalism and the World Trading System. Geneva, World Trade
Organization.

113. WTO. 2002 -— 4. Various online material accessible at htg)://www.wto.org
114. Yi, S.-S. 1996. “Endogenous Formation of Customs Unions under Imperfect

Competition: Open Regionalism is Good”, Journal of International Economics 41:
153-77.

183

 

 

   

I[iijgg‘nirijgij(it):I